Skip to main content

Full text of "CSAP Cultural Competence Series 6: Advanced Methodological Issues in Culturally Competent Evaluation for Substance Abuse Prevention"

See other formats


ompetence Series 

Advanced Methodolpgical 
l^ues irtPuftu rally 
Competent Evaluation 
for Substance Abuse 


Center for Substance Abuse Prevention 



Health Resources &: Services Administration 
Bureau of Primary Health Care 



CSAP Cultural Competence Series 6 

Advanced Methodological 
Issues in Culturally 
Competent Evaluation for 
Substance Abuse Prevention 

Offics of Minority Health 

R6scurc€s Center 

PO Box 37v337 

Washington, DG 20013-7337 

The primary objective of the Center for Substance Abuse Preven- 
tion (CSAP) Cultural Competence Series is to promote the devel- 
opment and dissemination of a scientific knowledge base that 
assists prevention program evaluators and practitioners in work- 
ing with multicultural communities. 

CSAP supports the rigorous evaluation of demonstration pro- 
grams designed to promote health and prevent substance abuse 
problems for all people. All positions taken on specific approaches 
to evaluating substance abuse problem prevention programs are 
positions of the communities, prevention experts, and authors who 
contributed to this monograph and may not necessarily reflect 
the opinions, official policy, or position of CSAP; the Substance 
Abuse and Mental Health Services Administration; the Health 
Resources and Services Administration; or the U.S. Department 
of Health and Human Services. Other groups that developed and/ 
or implemented specific methods for evaluating substance abuse 
prevention programs are documented in the text of this mono- 

All material in this volume, except quoted passages from copy- 
righted sources, is in the public domain and may be used or re- 
produced without permission from CSAP or the authors. Citation 
of the source is appreciated. 

DHHS Publication No. (SMA)96-3110 
Printed 1996 

Project Officer: 

Leonard G. Epstein, M.S.W. 

CSAP Cultural Competence Series: 

Elaine M. Johnson, Ph.D., Director, CSAP 

Ruth Sanchez-Way, Ph.D., Director, Division of State and 
Community Prevention Systems, CSAP 

Robert W. Denniston, Director, Division of Community 
Education, CSAP 


Ada-Helen Bayer, Ph.D. 
Senior Research Psychologist 
Fairfax, Virginia, and 
Assistant Professor 
Johns Hopkins University 
Baltimore, Maryland 

Frances Larry Brisbane, Ph.D. 


SUNY at Stony Brook 

School of Social Welfare 

Stony Brook, New York 

Amelie Ramirez, Dr.P.H. 


South Texas Health Research Center 

San Antonio, Texas, and 

Associate Professor 

University of Texas 

San Antonio, Texas 

Series Editor: 

Mario A. Orlandi, Ph.D., M.PH. 

Chief, Division of Health Promotion Research 

American Health Foundation 

New York, New York 

Managing Editor: 

Leonard G. Epstein, M.S.W. 

Division of Community Prevention and Training 

Center for Substance Abuse Prevention 

Substance Abuse and Mental Health Services Administration 

U.S. Department of Health and Human Services 

5600 Fishers Lane, Rockwall II 

Rockville, Maryland 20857 



Kenneth H. Cushner, Ph.D. 
Department of Education 
Kent State University 
Kent, Ohio 

Wei Li Fang, Ph.D. 
Office of Medical Education 
University of Virginia 
Charlottesville, Virginia 

Nancy Hepler, Ph.D. 
Department of Anthropology 
Memphis State University 
Memphis, Tennessee 

Barry Kibel, Ph.D. 

Pacific Institute for Research and Evaluation 

Chapel Hill, North Carolina 

Hyung-Chan Kim, Ph.D. 
Woodring College of Education 
Western Washington University 
Bellingham, Washington 

Sehwan Kim, Ph.D. 

"Just Say No'' International 

Oakland, California 

Walter J. Lonner, Ph.D. 
Center for Cross-Cultural Research 
Department of Psychology 
Western Washington University 
Bellingham, Washington 

Toshiaki Sasao, Ph.D. 

UCLA Department of Psychology 

University of California 

Los Angeles, California 


Joseph E. Trimble, Ph.D. 
Center for Cross-Cultural Research 
Department of Psychology 
Western Washington University 
Bellingham, Washington 

Robert T. Trotter II, Ph.D. 
Department of Anthropology 
Northern Arizona University 
Flagstaff, Arizona 

Philip B. Vander Velde, Ph.D. 
Woodring College of Education 
Western Washington University 
Bellingham, Washington 

Charles Williams, Ph.D. 
Department of Anthropology 
Memphis State University 
Memphis, Tennessee 


The Substance Abuse and Mental Health Services Administration's 
(SAMSHA) Center for Substance Abuse Prevention (CSAP) joins 
with the Health Resources and Services Administration's (HRSA) 
Bureau of Primary Health Care (BPHC) in producing Advanced 
Methodological Issues in Culturally Competent Evaluation for Substance 
Abuse Prevention, the sixth volume in a unique series of cultural 
competence publications. This volume explores questions of con- 
cern to evaluators who wish to perfect the art and science of work- 
ing with primary health care and substance abuse prevention 
programs serving different ethnic, racial, and cultural communi- 
ties. This volume is intended for substance abuse and primary 
health care providers, behavioral scientists, academicians, and 
students of the evaluation sciences who wish to broaden their 
expertise in the crucial issues that bridge culture and health. The 
topics addressed in this volume are pivotal to the growing inter- 
est of the managed care industry in fostering positive health out- 
comes, increasing the quality of and improving the accessibility 
to services, and increasing consumer satisfaction. 

The Cultural Competence Series has as one of its primary goals 
the scientific advancement of evaluation methodology designed 
specifically for substance abuse prevention approaches within the 
context of primary health care delivery in multicultural and eth- 
nically diverse community settings. Second, there is a strong need 
to expand the amount and quality of prevention studies that fo- 
cus on racial, ethnic, and culturally diverse populations; to build 
a knowledge base stemming from this research; and to improve 
primary health care program strategies. 

This volume, in the Cultural Competence Series, is a special 
joint collaborative effort between SAMHSA/CSAP and HRSA/ 
BPHC and fosters the mission of both organizations. Stimulating 
and supporting high-quality service delivery in prevention and 
primary health care to diverse populations are critical components 
of the mission of HRSA /BPHC. Applying research findings in 
order to strengthen and enhance substance abuse prevention pro- 
grams for diverse populations is a major responsibility of 


It is the sincere hope of those who have contributed to this 
series and to this volume that it will stimulate new ideas and pro- 
mote culturally appropriate health care and substance abuse pro- 
gram strategies for America's diverse communities. 

Elaine M. Johnson, Ph.D. 

Director, Center for Substance Abuse Prevention, 

Substance Abuse and Mental Health Services 


Nelba Chavez, Ph.D. 


Substance Abuse and Mental Health Services 


Marilyn H. Gaston, M.D. 

Assistant Surgeon General and Director, 

Bureau of Primary Health Care 

Health Resources and Services Administration 

Giro V. Sumaya, M.D., M.PH.T.M. 


Health Resources and Services Administration 



Advanced Methodological Issues in Culturally Competent Evaluation 
for Substance Abuse Prevention further defines and initiates the new 
discipline of culturally competent program evaluation. 

The first volume in this series. Cultural Competence for Evalua- 
tors: A Guide for Alcohol and Other Drug Abuse Practitioners Working 
With Ethnic/Racial Communities, explored the rich and diverse eth- 
nic heritage of the United States and the complex role that cul- 
tural factors can play in increasing either risk or resiliency. That 
monograph defined cultural competence as a set of academic and 
interpersonal skills that allow individuals to increase their un- 
derstanding and appreciation of cultural differences and similari- 
ties within, among, and between groups. This perspective requires 
a willingness to draw upon community-based values, traditions, 
and customs by working with individuals from and knowledge- 
able about ethnic/racial communities. The individual chapters in 
that monograph emphasized specific ethnic /racial populations — 
including African- American, Hispanic/Latino-American, Ameri- 
can-Indian, Alaska-Native, Asian- American, and Pacific Islander- 
American populations — and their respective substance abuse pre- 
vention profiles. 

Advanced Methodological Issues extends the efforts of the mono- 
graphs that preceded it by delving into more advanced topics and 
substantive areas relevant to program evaluation in diverse, cul- 
turally defined settings, regardless of the ethnic /racial popula- 
tions involved. Each chapter was developed from this broader, 
multicultural perspective and should be of interest, therefore, to 
all who want to advance their understanding of this highly com- 
plex topic. 

Advanced Methodological Issues is an outgrowth of a project that 
the Center for Substance Abuse Prevention (CSAP) undertook to 
explore the possibility of designing a multidisciplinary doctoral 
program in substance abuse prevention evaluation within an 
ethnic-specific and multicultural context. The curriculum for this 
doctoral program is available on the Internet (through PREVline, 
using this Telnet address: The file is available 

in WordPerfect 5.2 for Windows, filename CURRIC.WP (for tech- 
nical assistance call 1-800-729-6686). 

This work provides evaluators, academicians, researchers, and 
advanced students of culturally competent program evaluation 
with a clearly defined framework for learning more about this 
area, both through the monograph itself and through the continu- 
ing study it engenders. We hope that this monograph, and the 
series of which it is a part, will continue CSAP's celebration of the 
vast and rich diversity of the United States and of the unity within 
that diversity. We also hope that our readers will join us in this 

Mario A. Orlandi 
Leonard G. Epstein 


Foreword vii 

Preface ix 


Ada-Helen Bayer, Ph.D 1 

1 Acculturation. Ethnic Identificatioa and the 
Evaluation Process 

Joseph E. Trimble, Ph.D 13 

Social Change 13 

Social Conditions and Cultural Change 15 

Definitions of Acculturation and Ethnic Identity 18 

Acculturation and Ethnicity as Moderating Variables 27 

Measurement of Acculturation and Ethnic Identification 37 

Summary and Conclusions 52 

Acknowledgments 54 

References 54 

2 The History and Philosophy of Science in a 
Global Perspective 

Philip B. Vander Velde, and Hyung-Chan Kim, Ph.D 63 

Introduction 63 

Two Aspects of Science 64 

The Beginning of Science 65 

Cosmology and Science in Traditional China 66 

Social Science and the Newtonian Paradigm 70 

Newtonian Assumptions and Western Cosmology 71 

The Assumption Concerning Science and Schooling 73 

The Assumption Concerning Individuality 74 

The Assumption Concerning Nature 75 

The Assumption Concerning Anthropocentrism 75 

References 76 

Bibliography 77 

3 Psychonnetrics end Culture 

Walter]. Lonner, Ph.D 79 

Historical Considerations ,..,. 79 

A Major Focal Point: Etics and Ernies 81 

The Istanbul and Kingston Conferences 82 

Methodological Problems of Enduring Importance 83 

Types of Scales Used in Psychological Assessment 88 

Standardized Assessment 90 

Tests of Maximum Response 92 

Experimentation With a Possible Compromise: 

The SOMPA 97 

Tests of Typical Response 99 

Prototypical Examples of Measures of Typical Response ...... 100 

Response Sets 105 

A Variant of Measures of Typical Response: 

The Measurement of Attitudes, Beliefs, and Values 106 

The Measurement of Values 109 

Nonstandardized Assessment 109 

A Special Psychometric Case: Projective Methodology.......... 112 

Concluding Comments and Summary 114 

References 116 

4 The Role of Ethics in Evaluation Practice: 
Implications for a Multiethnocultural Setting 

Wei Li Fang, Ph.D 121 

Biomedical Model of Ethics 122 

Moral Issues ,.,.. 123 

Implications for Multiethnocultural Settings 126 

Ethical Dilemmas Encountered in Evaluation Practice 129 

Developmental Evaluation Model 150 

Summary 153 

Acknowledgments 153 

References 154 


5 Evaluating the Success of Community-Based 
Substance Abuse Prevention Efforts: Blending 
Old and New Approaches and Methods 

Sehwan Kim, Barry Kihel, Charles Williams, and Nancy Hepler ... 159 

Introduction 159 

Part I. Outcome-Focused Evaluation 161 

Part 11. Bridging to Research-Based Evaluation 169 

Part III. Some Closing Observations 178 

References 181 

6 The Cultural Context of Epidemiologic 

Toshiaki Sasao, Ph.D 183 

Abstract 183 

Introduction 184 

Purpose and Overview 184 

Understanding Ethnic-Cultural Diversity: 

A Double-Tiered Phenomenon 185 

Assumptions in Ethnic-Cultural Community Research 188 

Defining an Ethnic-Cultural Community 189 

Applicability of Cross-Cultural Research Methods and 

Concepts to Ethnic-Cultural Community Research 191 

Geographical or Ecological Instability of an 

Ethnic-Cultural Community 194 

Toward a Culturally Anchored Ecological Framework 

of Research in Ethnic-Cultural Communities 196 

References 207 

Author's Notes 212 

7 Culturally Specific Approaches to Knowing, 
Thinking. Perceiving, and Understanding 

Kenneth H. Cushner, Ph.D 213 

Defining Culture and Related Concepts 214 

Perception, Cognition, and Related Concepts 217 

Socialization and the Derivation of Meaning 218 

Western Versus Non- Western Conceptions of Self 220 

Culture-Specific World Views 223 


Subgroups in the United States: European- American 

Perspective , 226 

African-American Perspective .,,. 228 

Native- American Perspective ,. 230 

Asian-American Perspective 232 

Hispanic- American Perspective 235 

Cross-Cultural Interactions in the Delivery of Health 

and Counseling Services 237 

References 238 

8 Communication and Community 
Participation in Program Evaluation 

Robert T Trotter 11, Ph.D 241 

Introduction 241 

General Community Concepts and Issues 243 

Community Conditions That Affect Evaluation 245 

Stages of Community Involvement 248 

Use of Local Resources for Program Evaluation 253 

Communicating Evaluation and Program Research 

Findings ....262 

Conclusions and Recommendations 264 

References 265 



Ada-Helen Bayec Ph.D. 

Substance abuse prevention has been identified by the U.S. Gov- 
ernment as a major public health priority requiring intense and 
immediate attention. Further, we must improve our ability to 
evaluate the success and failure of programs targeting substance 
abuse prevention so that resources can be shifted to successful 
approaches, new programs can be developed using the best avail- 
able methods, and efforts can be concentrated on questions need- 
ing further research. In light of the racial and ethnic diversity found 
in our communities, it is of specific importance to conduct effec- 
tive and accurate evaluations in culturally diverse settings. An 
integral part of developing and enhancing skills in culturally com- 
petent program evaluation is to use the most methodologically 
favorable instruments and techniques available, and to develop 
evaluator's skills in understanding these methodologies within a 
variety of racial, ethnic, and cultural contexts. 

Advanced Methodological Issues in Culturally Competent Evalua- 
tion for Substance Abuse Prevention represents the sixth monograph 
in a series of cultural competence publications sponsored by the 
Center for Substance Abuse Prevention (CSAP). This monograph 
addresses and synthesizes the complex methodological issues in- 
volved in evaluation programs within multicultural contexts. 
Thus, the goal of this monograph is to integrate advanced meth- 
odological issues in program evaluation with cultural competence. 
Chapters in this monograph develop a framework and provide 
suggestions for evaluators who wish to use state-of-the-art meth- 
odological techniques to conduct culturally competent program 

This monograph consists of eight complementary chapters ad- 
dressing critical methodological issues in program evaluation 

within the culturally diverse settings found in our Nation. The 
following is a brief overview of the contents of each chapter. 

Chapter 1 

Joseph E. Trimble, in ''Acculturation, Ethnic Identification, and 
the Evaluation Process,'' begins by inviting the reader to consider 
the global changes that have occurred in the past two centuries, 
thereby engaging the reader in one of the fundamental issues of 
the monograph as a whole. Such global changes include environ- 
mental, social, imposed, and desired changes. The focus here is 
on human change, specifically in the sociological and psychologi- 
cal domains. Trimble reports that most changes in these domains 
are orderly and predictable; however, for some groups, for instance 
those that have been forced to immigrate to countries with very 
different cultural orientations, the change can be enormously dis- 
ruptive. Regarding such groups, discovering the conditions of 
change is especially difficult, yet necessary, work. According to 
Trimble, understanding why change occurs is an important as- 
pect of understanding the concepts of acculturation and ethnic 
identity because those concepts are not static psychosocial concepts. 

Definitions of the concepts are presented next. Any complete 
definition of acculturation must include the notion that it is a 
multifaceted, not unidirectional, form of sociocultural change in 
which moderating variables, individual preferences, and the de- 
sire for ethnic affiliation must be taken into account. Because ac- 
culturation affects each of the interacting groups involved, their 
ethnic identity is also affected. Defining ethnic identity is also a 
challenging task; Trimble cautions that care must be taken to avoid 
ethnic glosses and hierarchical nesting in identifying ethnicity. 

An informed overview of research themes concerning accul- 
turation and ethnicity as moderating variables in human thought 
and behavior is then provided. In reference to ethnic identity, the 
topics discussed are adoption, intermarriage, multiracial youth, 
personal adjustment, counseling, therapy, and substance abuse. 
Regarding acculturation, the topics described are adaptation pro- 
cess, acculturative stress, family functioning, counseling, drug and 
alcohol abuse, and successful adaptation. 

Measuring acculturation and ethnic identity yields some con- 
cerns, says Trimble; some researchers overlap the concepts, and 
many variables affect the concepts in similar ways. A multilevel 
approach of measuring acculturation is promoted, and the fluc- 
tuating nature of ethnicity must be recognized. Research 
on American Indian youths' alcohol involvement is highlighted. 
One purpose of the chapter is to stimulate thought about what 
culture itself is and therefore what it means to study and evaluate 
aspects of culture. 

Chapter 2 

Philip B. Vander Velde and Hyung-Chan Kim, in 'The History 
and Philosophy of Science in a Global Perspective,'' give an in- 
sightful account of how our perception of science relates to our 
perception of human beings and crises. The authors first explain 
why such an account must be considered for courses in doctoral 
programs in community psychology and related fields; second, 
they discuss the history of science in both the Western and East- 
ern cultures; third, they examine the main assumptions and re- 
sults of scientific thinking in those cultures. 

Community psychology is described as studying two usually 
separated concepts: the individual and the community Individuals 
are understood as embedded in communities, then the communities 
are understood as the central unit of analysis, and the moving dia- 
lectic of the two systems provides the grounding of the studies. 

The authors find that the history of science is a history of action 
and adventure; for instance, science requires both observation and 
imagination. If science is a way of understanding nature and the 
human being's relationship to it, then science began long ago, 
when classification and counting of objects in nature began. How- 
ever, a specifically scientific age is considered to have begun with 
the ancient Greeks. Western science originated from the notions 
of atoms and the four elements of nature and is characterized by 
its view that the universe is a series of independent and autono- 
mous events connected by a chain of causality. Chinese science, 
on the other hand, originated from the notions of the two forces 
and the five elements and is characterized by its view that all in 
nature is harmonious and mutually related. 

The view of nature and science from the Western perspective 
affects the Western view of social science, the authors contend. 
The Newtonian paradigm and Enlightenment views of rational- 
ity have created problems and yet are appealed to for solutions 
for those problems. Instead, the authors state, we must critically 
question the assumptions stemming from Newtonian physics. The 
primary assumption is that there is a knowable objective reality 
of isolated units; other assumptions involve science and reduc- 
tionistic schooling, abstracted individuality, control of nature, and 
anthropocentrism. For example, to understand human behaviors 
Western scientists work from the assumption that something 
within the individual, such as will, is the cause. Academic study 
should address these assumptions and attempt to discover the 
place and contexts in which human beings actually live. 

Chapter 3 

Walter J. Lonner, in 'Tsychometrics and Culture,'' views cross- 
cultural psychology as primarily a strategy in psychological re- 
search and treats the important issue of methodological problems. 
The psychometric method focused upon is the establishment of 
equivalence, and Lonner treats the main issues that stand in the 
way of the methodological aim: to minimize concerns about 
equivalence as a source of unwanted variance. 

After a brief review of historical considerations, the signifi- 
cant controversy regarding ''etics'' and "emics,'' and the milestone 
Istanbul and Kingston conferences, Lonner discusses the four 
types of equivalence (functional, conceptual, linguistic, and sca- 
lar or metric) that have been a concern for cultural studies and 
reviews the four types of measurement scales (nominal, ordinal, 
interval, and ratio) as used in psychological research. 

Lonner then examines three general areas of psychological as- 
sessment: behavioral, objective, and projective; each has its own 
character, as well as advantages and disadvantages. Objective tests 
can be divided into the categories of maximum response or typi- 
cal response. Tests of maximum response include those that at- 
tempt to determine a person's abilities, but because of the 
questions concerning the role culture plays in the actualization of 

human abilities, three approaches have been developed that ad- 
dress this question: the general abilities approach, the specific 
abilities approach, and the cognitive styles approach. Lonner dis- 
cusses the research and debate surrounding another approach, 
the System of Multicultural Pluralistic Assessment. In contrast to 
tests of maximum response, tests of typical response include at- 
tempts to determine attitudes and values. For such tests, examples 
are the Minnesota Multiphasic Personality Inventory and the 
California Psychological Inventory. Despite the etic-emic contro- 
versy, discussed previously in the chapter, people often agree that 
there exist several personological dimensions that are universal, 
such as anxiety, control, and depression, which are discovered 
using tests of typical response. In such testing, the problem of 
stylistic response must be dealt with. There is a large variety of 
tests to measure attitudes, beliefs, and values in variant tests of 
typical response; therefore, a look at some main considerations 
and techniques of such tests is important. Lickert scales, Thurstone 
scales, and Semantic Differential scales are three widely used scal- 
ing techniques; texts by Hofstede and Schwartz are recommended. 
Interesting nonstandard assessment procedures are the clinical 
method of Jean Piaget, cultural constructions, and cognitive maps. 
Projective methodology is not often used today in more sophisti- 
cated cross-cultural research for several reasons; it requires spe- 
cial training and caution to apply its techniques properly 

In essence, cross-cultural psychology and psychometrics can 
be joined, but several procedural problems must be overcome if 
the results are to be meaningful, just, and useful. 

Chapter 4 

Wei Li Fang, in "The Role of Ethics in Evaluation Practice: Implica- 
tions for a Multiethnocultural Setting,'' provides a sensitive account 
of the ethical issues involved with evaluation in multiethnocultural 
landscapes. The aims of the chapter are to examine the historical 
biomedical model of ethics, the moral issues related to evaluation 
process, the implications for multiethnocultural settings, the ethi- 
cal dilemmas encountered in evaluation practice, and to examine a 
possible ethical evaluation strategy and model. 

The historical biomedical model of ethics largely frames evalu- 
ators' views of ethics, says Fang. On this medical model, with its 
hierarchical form of moral reasoning, the patient is subject and 
the investigator is controller; full disclosure, confidentiality, and 
written consent are other characteristics. But significant differences 
exist between this model and the investigator-participant relation- 
ship of evaluation. 

Evaluation moral issues have received more attention from 
the profession than have program moral issues, but. Fang main- 
tains, evaluators should address both. The professional code for 
evaluation moral issues states that ethical decision making ought 
to be based on the principles of beneficence, respect, and justice. 
Program moral issues, such as public responsibility, must be ex- 
amined in reference to its principles as well. 

In the multiethnocultural setting, evaluators must be aware, 
sensitive, and appreciative of cultural diversity and of the impor- 
tance of meeting the needs of the various stakeholders. For in- 
stance, the evaluator must understand, as much as is possible, 
the problems and issues from the perspective of the targeted 
group. If cultural dimensions are not taken into account, misinter- 
pretation may result. The primary inclusion approach is suggested. 

Fang states that ethical dilemmas may be encountered in the 
evaluation practice when the evaluator and client are in conflict 
on ethical bases or in any stage of the evaluation process. Ethical 
dilemmas can arise in any of these stages: developing the evalua- 
tion contract; designing the evaluation; collecting, storing, and 
analyzing data; and interpreting and reporting evaluation find- 
ings. The role of the evaluator and training issues are also rel- 
evant. In developing the evaluation contract, several factors must 
be considered. A mutually negotiated contract takes into account 
seven concerns, according to Fang. In addition, when designing 
an evaluation, an ethically sound and culturally sensitive politi- 
cal process should be embraced. Propriety standards and organi- 
zational policies must be followed. Collecting, storing, and 
analyzing the program data should focus on the respect for indi- 
vidual and /or group autonomy and privacy. The stage of inter- 
preting and reporting evaluation findings ought to be presented 
in a useful, complete form. The evaluator 's role depends upon 

organizational context; problems differ for internal and external 
e valuators. Suggestions are made regarding training of evaluators. 
The chapter closes with a discussion of how the developmen- 
tal evaluation model is of significant value in the ethical making 
of evaluations for the multiethnocultural setting. 

Chapter 5 

Sehwan Kim, Barry Kibel, Charles Williams, and Nancy Hepler, 
in ''Evaluating the Success of Community-Based Substance Abuse 
Prevention Efforts: Blending Old and New Approaches and Meth- 
ods,'' address how to take the best from new and old types of 
evaluation, to result in the best method possible. The collabora- 
tive evaluator is described. The chapter consists of an Introduc- 
tion, followed by Part I: Outcome-Focused Evaluation, Part II: 
Bridging to Research-Based Evaluation, and Part III: Some Clos- 
ing Observations. 

The authors find that change on the institutional level is oc- 
curring in regard to substance abuse problem prevention activi- 
ties in the United States, due to necessity and funding issues. The 
principal force of such activities has been community and organi- 
zational coalition work. Such coalitions produce considerable chal- 
lenges, yet should be measured by their successes. Multiple 
intervention techniques have a greater chance for such success. 

Part I discusses the new views concerning evaluation. Evalu- 
ators in today's collaborative environments should supplement 
traditional evaluation ideas and practices with new ones, for four 
reasons. Basic tenets of good evaluation must apply, such as mar- 
rying content and methodology, and intimately involving the 
evaluator. The multiple-gates model is an application of these 
principles in coalition-based prevention; this model views transi- 
tions between stages as crucial points in the actualization of inter- 
vention. The transitions can be viewed as communicate-and-create 
movements. The roles for the evaluator of substance abuse prob- 
lem prevention are discussed. 

Part II addresses the ways in which the old views concerning 
evaluation must also be implemented. After identifying the criti- 
cal transition points or gates, applying the correct research 

methods to analyze the transitions requires traditional research 
tools. Here, cause-and-effect movement grounds the analyses. A 
hypothetical example shows how these methods might apply in 
a drug and alcohol abuse prevention program, and raises ques- 
tions for further study. Traditional techniques supplement, add a 
form of credibility to, and allow transferability of outcome-based 
evaluation. How to create control groups for intervention pro- 
grams is a problem, but eight helpful strategies that can some- 
times be used to compensate for the problem are presented. Six 
guidelines for designing evaluations in community-based, mul- 
tiple-intervention programs are also provided. 

Part III reopens the issue of change. Change does not happen 
in a vacuum, instead other parallel or at least coexisting changes 
occur: both an institutional and a private-sector management revo- 
lution are occurring. Five success principles of the management 
revolution are reviewed, which can be practiced in community- 
based coalitions as well. The outcome-focused evaluator, aided 
by traditional research methods, can foster such successful coalitions. 

Chapter 6 

Toshiaki Sasao, in ''The Cultural Context of Epidemiologic Re- 
search,'' expresses this purpose: to give a conceptual and meth- 
odological framework in which substance abuse researchers active 
in ethnic-cultural or multiethnic communities can make consci- 
entious decisions on conceptual and methodological issues based 
on a culturally anchored, ecological contextualist perspective. This 
broad, important goal is addressed in the following way: first, the 
definition of ethnic-cultural diversity is given; second, a critical 
appraisal of common assumptions is discussed; third, an appro- 
priate research framework is examined; and finally, the directions 
and limitations of the research are described. 

The definition of ethnic-cultural diversity is described as a 
double-tiered phenomenon of within-group and across- or be- 
tween-group diversity. Its complexity creates difficulties for re- 
searchers of ethnic-cultural communities; an important example 
concerns the etic-emic controversy 

Various methodological issues require resolution; however, 
more fundamental meta-methodological issues have not been 


given enough attention in the research of ethnic-cultural commu- 
nities, says Sasao. Community researchers, with the traditional 
focus on logical positivism, have been shortsighted in the knowl- 
edge of ethnic-cultural diversity and its implications for epide- 
miological study, and they have not met the needs of local 
ethnic-cultural communities. Therefore, three interrelated meta- 
methodological assumptions are addressed, regarding the con- 
cept of community, the applicability of cross-cultural theories, and 
the notion of stability 

According to Sasao, understanding an ethnic-cultural com- 
munity must also include an understanding of the larger context 
of the society where the target community is defined and embed- 
ded. So, moving toward the aim of a culturally anchored ecologi- 
cal framework of research in ethnic-cultural communities, the 
Cube Model is proposed and examined. The model is based on 
the concepts of cultural complexity and social representations, and 
typically generates three types of questions — epidemiological, 
etiological, and prevention /treatment intervention — which are 
reviewed. The notion of cultural complexity is explored at two 
levels (individual and collective) and three layers (acultural com- 
plexity, ethno-cultural complexity, and subcultural context). The 
model presented holds implications for further research, as well 
as limitations. Five directions for future research are suggested, 
and two limitations of the model are discussed. Sasao makes the 
final perceptive point that in using any appropriate model, hu- 
man judgment and experience are necessary to provide insight. 

Chapter 7 

Kenneth H. Cushner, in "Culturally Specific Approaches to Know- 
ing, Thinking, Perceiving, and Understanding,'' directs attention 
to the way in which understanding cultural views is helpful for 
delivering wellness services to cultural communities. Cushner 's 
aim is to provide professionals with broadened skills of analysis, 
interpretation, and communication. The chapter can be concep- 
tually divided into four parts: definitions of key terms, examina- 
tion of how human beings derive meaning from the world, 
exploration of different cultural conceptions of the self, and im- 

plications for cross-cultural interactions in the delivery of health 
and counseling services. 

Cushner begins by defining the relevant terms. Regarding the 
notion of culture, it is useful to distinguish its emic and etic as- 
pects. Regarding the notion of perception, psychologists are in 
some agreement about its sensory quality, but there is less agree- 
ment on the definition of cognition. 

Several views concerning the relationship between socializa- 
tion and the derivation of meaning are described. Multiple di- 
mensions contribute to how meaning is made of the world, and 
differing cultural experiences may end in very different cognitive 
styles. Furthermore, understanding notions of what knowledge 
is depends on consideration of the cultural origin of those notions. 

Conceptions of the self are influenced by culture, Cushner 
maintains. Therefore, the study of non-Western conceptions of self 
using Western theoretical constructs introduces the possibility that 
non- Western reality will not be understood. Numerous differences 
between Western and non- Western conceptions of the self are ex- 
amined, followed by a survey of international cultures and a brief 
summary of intranational subordinate groups. For these provi- 
sions, four dimensions of national culture and four subgroups of 
American culture are discussed. The European- American perspec- 
tive is the primary perspective of American culture; the subgroup 
perspectives include the African- American, Native- American, 
Asian- American, and Hispanic- American perspectives. Cushner 
sketches the subgroups' relatively recent historical experiences 
and analyzes how those experiences may influence perspectives, 
in the hope that health care and community personnel may better 
identify problems and solutions. Aware of the dangers associated 
with presenting cultural-specific accounts of groups, the author 
cautions that readers avoid stereotyping from these accounts. 

A discussion of cross-cultural interactions in the delivery of 
health and counseling services makes up the final section of the 
chapter. If provider and client have different backgrounds and 
culture-specific perceptions are not recognized, then providing 
services becomes problematic. Ilie author lists three considerations 
to be kept in mind when dealing with bicultural clients. Finally, 


Cushner emphasizes that illness behavior is understood and 
treated most effectively in its cultural context. 

Chapter 8 

Robert T. Trotter II, in ''Communication and Community Partici- 
pation in Program Evaluation Processes/' focuses on the signifi- 
cance of community involvement in culturally competent 
substance abuse programs. After some introductory remarks and 
general definitions, the conditions allowing for such community 
involvement are carefully analyzed and concluding remarks and 
recommendations are made. 

Trotter introduces the chapter by showing that the possibility 
of success or failure of culturally competent substance abuse pro- 
grams depends greatly upon the participation of local communi- 
ties. This dependence exists particularly within culturally 
competent substance abuse prevention programs. General com- 
munity concepts and issues are then defined. The concept of com- 
munity must be redefined for each target community and can be 
studied at various levels of interaction; the basic purpose of a com- 
munity is part of its definition. This community concept can help 
evaluators create the most appropriate environmental boundaries 
for their study. Trotter also describes what factors are involved in 
identifying a community unit. 

The main conditions that should be attended to in develop- 
ing successful programs and evaluation methods in culturally 
diverse communities are examined. One of the primary condi- 
tions concerns the cultural and sociopolitical environment in which 
the research is to be actualized. Local environmental values can 
have positive or negative impacts on substance abuse program 
results and on evaluation methods; Trotter identifies several is- 
sues regarding these environmental values. Another primary 
condition concerns the availability of local resources; local involve- 
ment is critical in the four phases of a program. A look at the 
types of local expertise usually available and at how to secure 
community cooperation for evaluation programs involves another 
condition, that of the utilization of local resources. The final 
primary condition Trotter describes is concerned with the 


communication of evaluation and program research findings. 
These valuable findings can face some common problems, and 
the researchers must take some impacts into account. 

Trotter concludes that working and thinking through the is- 
sues raised in the chapter requires an evolutionary, cyclical ap- 
proach. Dynamic policies, audiences, and relationships demand 
dynamic program elements. Moreover, seeing these issues from a 
cyclical perspective allows for the creation of culturally compe- 
tent evaluation programs; and it is just these kinds of programs. 
Trotter states, that may be much more acceptable to the commu- 
nities that substance abuse prevention programs hope to assist. 



Acculturation, Ethnic 

Identificotioa and tine 

Evaluation Process 

Joseph E. Trimble, Ph.D. 

Within the past 200 years the Western world has been experienc- 
ing vast changes. Every known element has been affected to the 
point that change itself is being challenged. At an environmental 
level we are witnessing the ravages of deforestation, the erosion 
of once abundant fertile land, the degradation of air quality, the 
depletion of natural food sources, and the consequences of mis- 
managed and poorly planned urban development. Changes are 
occurring rapidly also in the social structure and organization of 
Europe and the Americas. Environmental and social structural 
changes have produced corresponding changes in both indigenous 
and immigrant populations. For the indigenous aboriginal popu- 
lation of North and South America, changes invariably were im- 
posed through legislation, colonization, war, and disease. 
Immigrants expected change when they arrived; indeed, many 
immigrants from a wide variety of nations immigrated to the 
Western Hemisphere because they desired change. 

Social Change 

One undeniable, enduring characteristic of all humans, if not all 
living forms, is change. The study of acquired and imposed change 
forms the bulk of research initiatives in the social and behavioral 


sciences. At one level of inquiry, the interest is centered on social 
change that is the ''modification or alteration in the social struc- 
ture of a society" (Fisher, 1982, p. 488). One view of social change 
(Katz, 1974) spreads the forces of change among three discrete 
elements — from the individual to the structure of society to the 
psychosocial characteristics of group members. The three elements 
roughly follow the general level of inquiry in sociology, anthro- 
pology, and psychology. 

In sociology, presumably the progenitor of social change, top- 
ics such as social movements and collective behavior form the 
major source of information for change enthusiasts. In anthropol- 
ogy, the concepts of enculturation, acculturation, and assimila- 
tion include, wittingly or unwittingly, the concept of change 
brought about by societal influences. Typically, enculturation is ''the 
actual process of learning as it takes place in a specific culture" 
(Mead, 1963, p. 185), and acculturation is a form of cultural change 
that results from contact between two or more autonomous and 
independent groups; however, the change affects groups and their 
members in uneven ways (Berry, J., 1980). Assimilation often is a 
consequence of acculturation whereby an individual completely 
adopts and internalizes the values, beliefs, and behaviors of a 
dominant culture. In psychology, change, especially at the indi- 
vidual level of analysis, is explored through studies on learning, 
clinical intervention, individuation, socialization, conformity, 
modernization, and, to a lesser extent, acculturative stress and 
identity formation. This chapter focuses on the sociological and 
psychological domains, along with the broader concepts of accul- 
turation and ethnic identity. 

To appreciate the significance and importance of the two con- 
cepts, it is necessary to provide some background on their origins 
and subsequent growth in the behavioral and social sciences. Rec- 
ognize that acculturation and ethnic identity are not static 
psychosocial concepts. Although ethnic identity may be more 
stable and enduring, change, nonetheless, is an integral part of 
the current environment; change is thus an important concept in 
understanding acculturative processes in humans. Therefore, the 
topic of change and its sociocultural correlates will be discussed 


in more detail. Following this section, definitional and historical 
background information will be presented. Discussion about con- 
cepts similar to acculturation will follow and lead to a discussion 
of a few variables thought to mediate acculturation and identity. 
Material specific to ethnic /racial populations will then be pre- 
sented to highlight and emphasize the way each group responds 
to conditions that influence both acculturation and ethnic iden- 
tity The chapter ends with a section devoted to measurement top- 
ics. The overall intent of the chapter, consequently, is to present 
fundamental background material to both provide a foundation 
and promote further reading and research. 

Social Conditions and Cultural 

A number of behavioral and social scientists take the position that: 
Social change is accompanied by the intensification of social and 
cultural sources of psychological conflict by new stresses and 
new adaptation requirements in new milieus, and by the loss of 
the stabilizing effect of old cultural patterns. (Kiev, 1972, p. 9) 

Much of the social science research tends to concentrate on 
the consequences and impacts of social change agents on indi- 
viduals, families, small communities, and even large cultural and 
religious groups. Human adaptation and coping with sudden 
changes and life crises have formed the theme of a collection of 
articles devoted to various forms of situational and personal ac- 
tions that evoke sudden change (Moos, 1976). 

All changes require use of coping strategies. Sudden and very 
rapid social changes, however, are likely to be especially disrup- 
tive for individuals. The abruptness of a change, some argue, is a 
central source of individual stress (Kiev, 1972; Marris, 1974) lead- 
ing, in many cases, to psychiatric illness, shock, and death. Others 
argue that the sudden change produced by the coming together 
of different cultural traditions produces ''acculturative stress" 
(Berry, J., 1980) leading to the use of such destructive coping 
mechanisms as substance abuse at the individual level and 

massive disruption of cultural values and norms at the group level. 
Abrupt change can be disruptive especially when individuals and 
groups are not totally informed of the conditions and the conse- 
quences produced by the change. Under these conditions, indi- 
viduals may demonstrate the inability to produce or evolve coping 
mechanisms to deal effectively with the strains and stresses of 
change. Efforts to utilize known defense, coping, and adaptive 
mechanisms seem to become dysfunctional in situations created 
by the abrupt presence of change. Unfortunately, recognition of 
the ineffectiveness and inappropriateness of mechanisms does not 
come about during or immediately after a sudden change. Instead, 
it typically occurs after a vicarious effort to use known or socially 
acceptable mechanisms to deal with the new conditions created 
by the change. As each effort to deal with strange and unfamiliar 
situations meets with failure, stress, anxiety, and destructive forms 
of behavior seem to increase. Many persons, through self-initiated 
efforts or intervention by others, manage to pull out of the crisis. 
Many are strengthened by the experience. Others succumb to their 
never-ending search for structure and meaning created by the new 
circumstances. Individual responses to rapid change are some- 
what like the effect of fire on material objects — some are consumed, 
and some are strengthened. 

Although personal coping mechanisms are one strategy, so- 
cial change can also be dealt with through adaptation, mastery, 
and defense (White, 1974). Each form of adjustment requires a 
particular kind of behavior typically learned within one's cultural 
environment as a result of past experiences and shared experi- 
ences from one's group members. 

Stress and Change 

Stress is another dramatic response to the circumstances created 
by social change. However, stress reactions can be intimately 
linked to coping and adaptation efforts. A stressful experience 
may be sparked by a generalized stressful situation and will lead 
to attempts to cope effectively with the circumstances. However, 
if coping efforts fail the stressful experience can be intensified. 
Viewed in this manner, a systematic relationship exists between 


change, coping, and stress and, as suggested by Manderscheid, 
Silbergeld, and Dager (1975), this relationship can generate nega- 
tive cybernetic feedback loops. 

As suggested by Leighton (1949), basic stress types, such as 
threats to life and health, loss of means of subsistence, enforced 
idleness, restriction of movement, and capricious and unpredict- 
able behavior on the part of those in authority upon whom one's 
welfare depends can create additional disturbances, including 
frustration of expectations, ambivalence, suspiciousness, hatred, 
hostility, destructive action, and withdrawal and resignation. 

Smelser (1968) suggests that the above can constitute inde- 
pendent variables, where the reactions to stress form the unit of 
analysis, and per Leighton's suggestion result in patterns of be- 
havior that may consist of the following: 

• Constructive activity directed toward overcoming the 
source of stress 

• Vicarious activity evoked in succession 

• Suspiciousness, hatred, hostility, and destructive action 
sometimes redirected toward surrogate causes instead of 
the actual ones 

• Withdrawal and resignation leading to a state of hope- 
lessness and futility 

Certainly, these four response patterns are but a few of the many 
reactions that can occur at a psychological and physiobiochemical 
level. Other factors are known to relate closely with stress reactions. 
Studies over the past two decades have led investigators to con- 
clude that minority and social status (Dohrenwend, 1973; Vaughn, 
Lin, & Kuo, 1974) and age and sex (Dupuy Engel, Devine, Scanlon, 
& Querec, 1970) are strong predictors of stress vulnerability. Simi- 
larly, some investigators have isolated certain personality character- 
istics as potent contributors to stress reactions (Coelho, Hamburg, & 
Adams, 1974). Thus, each person's vulnerability to stress and the 
availability of responses to stress affect the reactive style of individuals 
experiencing stress and, indeed, stem from one's own cultural mi- 
lieu and experiencing background. 


Definitions of Accuiturction and 
Etinnic Identity 

The Social Science Research Council (SSRC) defined acculturation 
as ''culture change that is initiated by the conjunction of two or 
more autonomous cultural systems. Its dynamics can be seen as 
the selective adaptation of value systems, the processes of inte- 
gration and differentiation, the generation of developmental se- 
quences, and the operation of role determinants and personality 
factors" (SSRC, 1954, p. 974). The key concepts in the SSRC defi- 
nition are change and adaptation. Subsequent research and ex- 
ploration of the two processes generated different views of the 
acculturation construct. 

The more traditional definition implies that a cultural group 
moves from a native- or tradition-oriented state through a transi- 
tional stage and progresses to where one reaches an ''elite accul- 
turated" stage (Spindler & Spindler, 1967). According to this 
notion, cultural changes proceed away from one's own cultural 
lifeway in a linear manner and culminate in the full and com- 
plete internalization of another culture's lifeway. More contem- 
porary social researchers have difficulty with the traditional view 
claiming that acculturation is neither a linear process nor an 
achievable end, especially if the process occurs during the initial 
contact and change period. Many social scientists, in fact, would 
argue that no groups or individuals are fully acculturated if some 
vestige or relic of their traditional culture remains with them. If 
the elements of a donor have not been fully and thoroughly in- 
ternalized, then full acculturation cannot occur; if it does, it may 
take several generations for the process to become complete. Some 
groups choose to select portions of a donor 's culture that fit with 
their world view and, at the same time, strive to retain vestiges 
of their traditional culture. 

The small number of researchers promoting and advancing 
acculturation research are adopting bidimensional and multidi- 
mensional perspectives. Bidirectional perspectives view accul- 
turation as a process in which elements of both their own and 
donor culture are retained and internalized (see Le Vine & Padilla, 


1980). Mendoza (1984) and Sodowsky, Lai, and Plake (1991) view 
the construct from a multidimensional perspective. Instead of at- 
tempting to isolate an individual on an index that approaches full 
assimilation, one must consider the possibility that many options 
are available and depend on the situation. Mendoza suggests that 
an acculturating individual may reject religious practices, assimi- 
late dress customs, and integrate food preferences and the cel- 
ebration of certain holidays — one's acculturative status, therefore, 
is best understood from a composite of indices rather than from 
an aggregated summative index. Trimble (1988b) also advocates 
a similar view and emphasizes the intricate recursive nature be- 
tween and among person variables, situational settings, and 
acculturative patterns — the model emphasizes the potency of con- 
textual and situational variables in determining behavior, percep- 
tion, and cognitive appraisals. 

The acculturative process was once thought to be a unidirec- 
tional course of cultural change eventually resulting in full as- 
similation. There is growing evidence that suggests that 
acculturation is more multifaceted, however Moderating vari- 
ables, individual preferences, and the desire for ethnic affiliation 
must be factored into the process. When immigrant groups con- 
verge and interact with one another, changes are inevitable. Even 
the host culture will experience change. But along with the changes 
in lifestyle come changes and challenges to ethnic identity. 

Defining Etiinic Identity 

Just as there are several definitions for acculturation, a construct 
like ethnic identity generates many viewpoints. To understand the 
complications one must consider the meanings oirace dind ethnicity. 
Feagin (1978) defines a racial group as one in which "persons in- 
side or outside the group have decided what is important to single 
out as inferior or superior, typically on the basis of real or alleged 
physical characteristics subjectively selected'' (p. 7). An ethnic 
group, maintains Feagin, is one that "is socially distinguished or 
set apart, by others and /or by itself, primarily on the basis of cul- 
tural or nationality characteristics" (p. 9). Thompson (1989) elabo- 
rates on the term ethnic and chooses to view an ethnic group as a 


culturally distinct population that can be set apart from other 
groups. Such groups, Thompson argues, engage in behaviors 
''based on cultural or physical criteria in a social context in which 
these criteria are relevant'' (p. 11). 

Helms (1990) maintains that racial identity actually refers to a 
sense of group or collective identity based on one's perception that 
one shares a common racial heritage with a particular racial group. 
"Ethnic" and "race" often are used interchangeably — even "cul- 
tural" is tossed in occasionally The three terms may share a com- 
mon meaning, but only from the perspective that people 
congregate around common core characteristics. The source of the 
core characteristics can be criteria established and deeply held by 
the in-group; out-groups, too, can set their own criteria for desig- 
nating and differentiating one group from another. 

To distinguish one group or individual from another by appeal- 
ing to race, ethnicity, or culture is an attempt to make distinctions. 
Labeling a group as a distinct cultural (racial, or ethnic) unit, how- 
ever, tends to promote stereotyping and leads to overgeneralizations 
further compounding the complexity of the problem. It is not un- 
common for outsiders to believe that identifiable members of a ra- 
cial group act as a single unitary whole — a group mind — and that 
they are more homogeneous than heterogeneous. Such labeling leads 
to blanket statements and stereotyping. 

Numerous debates are emerging in the literature that chal- 
lenge many of the theories espoused to explain the complex mean- 
ing and nature of ethnic identity. Richard Thompson (1989) 
provides a controversial perspective on the topic. Thompson takes 
on the primordialist position — a viewpoint that is used to explain 
and justify emerging "ethnic consciousness" movements — ^by ap- 
pealing to other more seemingly social structural perspectives such 
as assimilationism, capitalism, and neo-Marxism. He also lays out 
a sequence of interesting topics that any theory of ethnicity should 
accommodate: (1) ethnic and racial classifications; (2) ethnic and 
racial sentiments; and (3) ethnic and racial social organizations. 
Humans typically develop systems to categorize and classify 
themselves and others, attach significance and meaning to the 
classification, and use racial and ethnic classifications for organi- 
zational criteria. 


Attempts to define ethnicity and ethnic identification are no 
less complicated than attempts to define race and culture. At one 
time, anthropologists generally agreed that humankind could be 
grouped into four racial categories, specifically Australoid, 
Caucasoid, Negroid, and Mongoloid. The physiognomic charac- 
teristics of each category were used as primary criteria for distin- 
guishing one racial group from another; some anthropologists 
went a bit further to consider use of common yet distinctive gene 
pools. Today, however, the concept of race has fallen from use in 
the anthropological literature, all the more so because the once 
isolated gene pools are disappearing, especially in the Western 
Hemisphere. There are reasons for abandoning race as a classifi- 
cation, especially from a psychological perspective: simply put, 
individual personality and behavior differences cannot be attrib- 
uted to gene pools. Use of the term racial identity is anachronistic, 
psychologically and politically charged, and has little value in 
explaining ''within group'' variations. 

Some attention also should be given to the term cultural iden- 
tity. It, too, is a complex term and is probably the most difficult of 
the three to define. There are an array of complications owing to 
theoretical differences; however, the definition oi culture (cultural) 
presents the biggest problem. Anthropologists and cross-cultural 
psychologists claim that there are over 120 different definitions 
of culture (Kroeber & Kluckhohn, 1952). In defining culture one 
must consider not only what exists, but also what is desirable. 
Cultures are not merely an amalgam of specific behaviors because 
they include activities of a standard nature: consistent cognitive, 
perceptual, motivational, and affective patterns, and a distinc- 
tive array of artifacts of human alterations of the environment. 
Cultures also are continuous, cumulative, and progressive (White, 
1947). And culture indeed is an abstraction. Thus, it can be a way 
of life attributed to a distinct collective of humans that reside in a 
particular geographic locale; however broad this definition may 
be, it is not without flaws. So, for the sake of brevity, we can ap- 
peal to Herskovits' (1948, p. 7) definition; he maintains that "cul- 
ture is the man-made part of the human environment." 
Primordialists and sociobiologists undoubtedly would have some- 
thing to say about this concise and somewhat simplistic version. 


Nonetheless, the reader is encouraged to review the numerous 
definitions of culture. 

The "Ethnic Gloss" 

Before the topic of ethnic identity is considered in more detail, 
attention must be given to the manner in which researchers, among 
many others, specify and describe ethnic and culturally distinct 
populations. As one scans the ethnic /racial population mental 
health and substance use literature, it becomes readily apparent 
that there are a number of studies that focus on American Indians 
(or Native Americans), Asian Americans and Pacific Islanders, 
African Americans, Mexican Americans and Puerto Ricans, and 
other ethnic-specific groups. Occasionally, researchers provide 
specificity concerning their respondents in their titles and abstracts 
by referring to a geographic region or a city in the United States, 
for example, the lower Rio Grande Valley, Seattle, Los Angeles, or 
the Southwest. Others will distinguish their respondents along 
urban and rural lines, while others, when referring to an Ameri- 
can Indian group, will specify the tribe (for example, Mohave, 
Chippewa, or one of the Pueblos). For a vast majority of the stud- 
ies in the ethnic /racial population literature, descriptions of eth- 
nic and cultural groups tend to rely on the use of broad ethnic 
glosses. Basically, an ethnic gloss is a superficial term that is used 
to refer broadly to a specific group without giving any attention 
to the heterogeneity that exists within the group. Use of such 
glosses gives little or no sense of the richness and cultural varia- 
tion within these groups, much less the existence of numerous 
subgroups characterized by distinct lifeways (also known as ethos) 
and thoughtways (also known as eidos). Furthermore, the use of 
broad ethnic glosses to describe a group in a research venture may 
be poor science. Apart from the fact that such sweeping references 
to ethnic groups are gross misrepresentations, their use can vio- 
late certain tenets concerning external validity and hinder the 
ability to generalize findings across subgroups within an ethnic 
category — they erode any likelihood of an accurate and efficient 
replication of research results. 


The use of ethnic glosses to describe subjects and respondents 
has generated many concerns in recent years. Critics point to the 
fact that the major ethnic groups (specifically American Indians 
and Alaska Natives, Asian Americans and Pacific Islanders, 
African Americans, and Hispanics) represent varied sociocultural 
and subgroup categories. Describing "American Indians,'' a 
widely used and abused ethnic gloss, Trimble and Fleming (1989) 
point out that American Indians represent an extremely diverse 
and complicated ethnic group. There are well over 450 identifi- 
able tribal units, in which individual members represent varying 
degrees of mixtures resulting from intermarriages and reflect vary- 
ing acculturative orientations that affect ethnic identity. Wong 
(1982), commenting on the labels ''Asian Americans" and "Pa- 
cific Islanders," asserts that at least 32 distinct ethnic and cultural 
groups might meaningfully be listed under these designations and 
that the differences among and between these groups are extraor- 
dinarily complex. Morishima, Sue, Teng, Zane, and Cram (1979) 
add that "given the diversity of languages, norms, mores and 
immigrant/ American bom [status], it is evident that to [label these 
peoples as Asian American and Pacific Americans] implies a ho- 
mogeneity which is lacking" (p. 3). And writing about the "His- 
panic" gloss, Padilla and Salgado de Snyder (1985) state 
emphatically that Hispanic "is a term used to designate those in- 
dividuals who reside in the United States and whose cultural ori- 
gins are Mexico, Puerto Rico, Cuba, and other Latin American 
countries." As such, the term Hispanic is not accepted by many 
individuals, and it is not uncommon to find reference to Latinos 
or La Raza in place of Hispanic in some communities. 

In selecting ethnic samples for social and behavioral science 
studies, researchers almost tacitly assume that the respondents 
share a common understanding of their own ethnicity and na- 
tionalistic identification. It is as though the researcher believes 
that American Indians, African Americans, and others share some 
modal characteristic that at one level sets them apart from an- 
other comparative sample such as "Whites" (Trimble, 1988a). 
There is mounting evidence, though, that suggests that the 


assumption maybe completely mistaken. Marin and Marin (1982) 
illustrate how the use of an ethnic category, in this case Hispanic, 
can disrupt a well-intended sampling strategy. The researchers 
were interested in the health records of some 500 patients at a 
clinic in East Los Angeles, California, an area known for its high 
concentration of people with Hispanic origins. Because of the lo- 
cation of the clinic they expected to find mostly Hispanics in their 
sample. Much to their surprise, they found that some patients 
with Spanish surnames had actually checked the "White" or 
"other" ethnic category on the medical form. Upon examining 
the "other" category, they found that 3 percent of the patients had 
written in specific ethnic identifiers such as "Mexican- American" 
and "Chicano." All in all, some 13 percent chose to identify them- 
selves in a way that differed from what one might expect from a 
person with a Spanish surname. Indeed, the use of "Hispanic" as 
a means to identify a sample is insufficient since the category can 
mean quite different things to different people. 

Broad-brush ethnic categories do have a useful purpose at one 
level. Researchers and policy planners can use the results to dif- 
ferentiate groups to highlight salient characteristics. In a crude 
way, it is a good deal easier to discuss differences between ethnic 
and cultural groups by using the gloss (Trimble, 1991). Thus, plan- 
ners, legislators, and decision makers find it convenient to bal- 
ance and present the findings and problems of one or more ethnic 
groups against other groups. One may consider the use of the 
ethnic gloss at this level to be tolerable. 

Most definitions of ethnic identity rely on an identifiable group 
to form the basis of the construct. Reliance on a group orientation 
not only views identity as static but also provides little credence 
to individual level variations and influences created by situational 
or contextual circumstances. In short, psychological and situational 
characteristics are not found in typical definitions and 
conceptualizations of ethnic identity Thoughts on this point are 
offered in the following section. 


Ethnic Identity and l-iierarctiical Nesting 

Ethnic groups can be viewed as units of analysis. To identify and 
discover the ethnic perspective of a unit, one should first con- 
sider compiling an inventory (see Berry, 1985). The investigator 
should carefully explore the concept of a group's ideas about eth- 
nic identification. Determining one's ethnicity provides research- 
ers with data that enable them to establish more compact 
homogeneous subgroups. For example, identifying value orien- 
tations provides data that reveal the extent to which a respondent 
endorses traditional indigenous values against those more repre- 
sentative of a dominant culture or the one serving as the major 
acculturating agent. Presumably, a native-oriented, nonacculturated 
individual would endorse and act out native-oriented values, and 
one who expressed the least ethnic identification would espouse 
values more in line with the group with which he or she affiliates. 
Once data are accumulated, researchers can then proceed to con- 
duct ethnic-group-specific and comparative cross-cultural stud- 
ies by introducing other variables of interest. 

One type of respondent identification procedure, hierarchical 
nesting, results in a rather subjective interpretation of one's 
ethnicity and value orientation. In fact, the method comes close 
to the main premises of symbolic interactionism, which are that 
"(1) humans act toward things on the basis of the meanings that 
the things have for them; (2) the meaning of the things is derived 
from, or arises out of, the social interaction that one has with one's 
fellows; and (3) the meanings are handled in, and modified 
through, an interpretative process used by the person in dealing 
with the things he encounters" (Blumer, 1969, p. 2). In more spe- 
cific terms, one's ethnicity and expression of values are central- 
ized meanings particular to the individual and, therefore, have 
intrinsic importance in their own right. 

It can be argued that ethnic identification is a subjective expe- 
rience designed "to express affiliation, allegiance or oneness" with 
a preferred group (Casino, n.d., p. 16). Equally important is the 
notion that ethnic identity is contextual and that it "is a product 
of social transaction [my emphasis] insofar as one assumes an ethnic 


identity by claiming it and demonstrating the conventional signs 
of membership. A claimant is always subject to the response of 
others who may concur with or deny the claim'' (Casino, p. 18). 

Efforts at establishing ethnic identity typically occur at an 
exonymic level (Casino, n.d.), where the researcher presents a set 
of fixed-attitude-like statements to the respondents. Often the eth- 
nic identity scale resembles a pseudo-etic set of items. What 
emerges is an outsider-produced level of identity, the exonymic, 
not the respondent's autonymic, definition. Literally, exonymic 
means ''outside names or labels" and autonymic means "self 
names or labels." Ethnic actors indeed embody an ethnic con- 
sciousness (Klineberg & Zavalloni, 1969) that is closely aligned 
with the unique ethos and eidos of the group they affiliate with. 
The ultimate test, of course, is "the authentic union of personal 
identity (the autonym) with communal identity" (Casino, p. 17). 
Therefore, it is logical to assume that a concordance would exist 
between the autonymic and the exonymic where the importance 
is placed on the individual's own categories and intentions for 
self-identification. Yet it would be foolhardy to assume that one's 
criteria would not line up with those of the group — often, how- 
ever, the criteria developed by researchers do not align with the 
group's criteria and hence may well be conceived as pseudo- 
exonymic or, worse yet, an ethnic gloss. 

Neoidiographic ethnic identification involves more than 
merely nesting oneself in a hierarchical scheme. More often than 
not, individuals will use rather ethnolocal speech patterns and ges- 
tures to promote the authenticity of their claim. If outward physi- 
cal appearances do not mesh or there is the sense that the other 
party doubts the identity claim, ethnic actors will tend to exag- 
gerate and give emphasis to mannerisms and speech 
idiosyncracies known to be particular (or peculiar) to the group 
in question. The somewhat stylized ritual can be referred to as 
situational ethnicity whereby ethnic actors take the occasion to re- 
affirm their ethnicity, often to the dismay and puzzlement of out- 
siders. Including these ethnic-specific mannerisms in a measure 
of ethnic identification, though a worthy research effort in its own 
right, would be awkward, time consuming, and possibly redundant. 


Acculturation and Ethnicity 
as iVIoderating Variables 

In the past few decades, a number of researchers have been ex- 
ploring the relationship between a variety of social and psycho- 
logical variables and processes that are related to or influenced 
by the acculturative process and ethnic identification. Much of 
the work focuses on social and individual change; some of the 
research describes certain characteristics presumed to be related 
to identity and acculturation. In general, the research themes are 
variegated and follow no identifiable thrust or theme; instead, 
apart from a few studies, research findings are scattered among a 
variety of topics. To provide an overview of the themes, exem- 
plary studies are presented organized around topical themes for 
each of the two constructs of ethnic identity and acculturation. 

Ethnic Identify 

Adoption, Internnarriage, and Multiracial youth 

A number of studies point to a potential relationship between 
intermarriage and adoption and degrees of ethnic identity both 
among married partners, offspring, and added family members. 
Judd (1990) found that in children from families where parents 
differed in religious affiliations, material influences can serve to 
maintain or enhance the child's religious-ethnic identity, especially 
if the child follows the mother's religious preference. In a related 
study, Parsonson (1987) found that children's ethnic identities are 
stronger when parents come from the same ethnic group. 

Socialization experiences of mixed race and transracial 
adoptees can influence ethnic identity and awareness of one's 
own natal culture. McRoy, Zurcher, Lauderdale, and Anderson 
(1984) found that the degree to which families discussed racial 
identity issues can affect transracial youths' perception of them- 
selves when compared to members of their own ethnic group; 
hence, identities varied as a function of the cultural composition 
of a family's network. Andujo (1988) also found that socioeco- 
nomic factors can influence an adopted child's ethnic identity. 


Personal Adjustment Counseling, and Therapy 

Immigration and relocation are potent forms of sociocultural 
change. A few researchers have conducted studies to examine the 
effects of resettlement on ethnic identity, especially among Asian 
populations. Bromley (1988) draws attention to the identity for- 
mation problems occurring among Southeast Asian adolescent 
refugees, especially those who came unaccompanied to the United 
States. She points out that four distinct conditions affect identity 
development among this age group: exposure to overwhelming 
tension-producing situations; failure in the practice of proper val- 
ues; withdrawal from channels that promote identity formation; 
and inadequate coping mechanisms. 

Forming a solid identity with one's own ethnic group can be 
accompanied by some behavioral and personal problems. 
Grossman, Wirt, and Davids (1985) found among samples of 
Chicano and non-Chicano White youth that ethnic esteem medi- 
ates with self-esteem. Chicano youth expressed strong ethnic es- 
teem but demonstrated low self-esteem and low behavioral 
adjustment patterns. Gonzalez (1988), in a related study, found 
that high-achieving Chicano women who expressed strong eth- 
nic esteem tended to be a threat to Chicano males; the threat per- 
ception tended to create distress for the women. Maintenance of 
ethnic identity, therefore, created a conflict between achievement 
and the desire to maintain close interpersonal and cultural rela- 
tions with men. 

Ethnic identity often is introduced as a presenting problem in 
counseling and therapeutic settings. Ruiz (1990) introduced a 
model that captures Chicano /Latino identity problems that can 
occur in a counseling setting; recognition of the various identity 
stages can assist the counselor to propose varied interventions to 
assist the troubled client. Similarly, Comas-Diaz and Jacobsen 
(1987) argue that members of varied ethnic/racial groups not only 
experience problems with their ethnocultural identities, but of- 
ten attribute certain ethnocultural qualities to their therapists; the 
attribution process can be turned into an effective therapeutic tool. 

To counter some of the problems encountered by ethnic-unique 
clients, many practitioners emphasize the importance of appro- 
priate client- therapist matching. Matching clients with therapists 


with similar ethnocultural orientations presumably facilitates cli- 
ent responsiveness and counselor effectiveness. Yet in a review of 
the literature on this recommendation, Flaskerud (1990) points 
out that client- therapist matching may not truly facilitate the thera- 
peutic process. Helms (1989) in part agrees and goes on to argue 
that findings from Black racial identity theory research may con- 
found the plausibility of applying and integrating results with 
counseling psychology approaches. 

Psychological Nigrescense, Identity, and Self-Esteem 

In recent years, a good deal of attention has been devoted to the 
salience of ethnic identity for African Americans. Mays (1986), in 
a compelling article, describes the various historical forces that 
shaped the identity of Blacks. Using a psychological framework, 
she emphasizes the influences that various social and economic 
forces have had on an ethnic population that was essentially pow- 
erless to effect change. Ethnic and group identity among Blacks 
was strong and robust. The strength of that identity was supported 
in the results of a national survey of Black Americans by Broman, 
Neighbors, and Jackson (1988), in which the older and least edu- 
cated Blacks expressed the strongest identity 

The literature on African- American identity quite often sug- 
gests that some African Americans hold differing opinions about 
who and what they are. Parham and Helms (1985) and Parham 
(1989) explored the process of psychological nigrescence (the pro- 
cess of identifying oneself and becoming Black) among different 
Black populations in an effort to support and extend the Cross 
Model of Psychological Nigrescence. In one study, the authors 
found that pro-White /anti-Black and pro-Black/ anti- White stu- 
dent attitudes were associated with personal distress. Awaken- 
ing Black attitudes were positively associated with the process of 
self-actualization and negatively associated with inferiority feel- 
ings and anxiety levels. Their findings hold some importance for 
cross-cultural counselors. 

Undoubtedly, identity is a major portion of what it means to 
be an African American in a multicultural society Looney (1988) 
demonstrated that an interesting relationship existed between ego 
development and Black identity: if one has a strong ego then one 


will define who and what one is; however, if the ego is weak then 
one will allow others to do the defining. Identity can also be in- 
fluenced by the sociocultural context in which one is reared, so- 
cialized, and educated. Baldwin, Duncan, and Bell (1987) found 
that Black identity was higher among Black students who attended 
a predominantly Black university than among Blacks who at- 
tended a predominantly White school. Therefore, the context in 
which one finds oneself reifies one's sense of belongingness and 
one's identity. Also, the extent to which socioeconomic status re- 
lates to ethnic identity among African Americans is questioned 
(Carter & Helms, 1988). Finally, it should be noted that the explo- 
ration and meaningfulness of ethnic identity appears to be more 
important for ethnic /racial population youths than for White 
youths (Phinney & Alipuria, 1990). 

Substance Abuse and Ethnic Identity 

Westermeyer (1984) argues that the relationship between ethnic 
identification and substance abuse patterns is not straightforward. 
In certain settings, users form subcultures for which substance 
abuse is the central organizing factor whereby the meaning and 
patterns of such abuse serve to reify identification with the group 
(Moore, 1990). To the contrary, for some ethnic /racial population 
youths, substance abuse is a way to deal with ethnic identity prob- 
lems, especially the "hatred'' for being a visible member of the 
group (Galan, 1988). Still, for other ethnic-specific substance abus- 
ers, the abuse rates may serve as a protest against the majority 
culture and the oppressive experiences created by discrimination 
and racism. Because of the complexities surrounding the possible 
relationship between ethnicity and drug abuse, researchers have 
shied away from vigorously pursuing the question; there are very 
few studies in the reported literature that address the relationship. 


From all appearances, the research literature of the past decade or 
so suggests that researchers believe that acculturative status con- 
tributes to a wide range of sociocultural and behavioral problems. 
Since the concept of acculturative status implies that culture con- 


tact produces social and psychological changes, one would ex- 
pect to find such topics as acculturative stress, counseling readi- 
ness and acceptance, familism and ethnically mixed marriages, 
and drug and alcohol use in the literature. And that is the case. 
Along with these topics, researchers have been struggling with 
the development of measurement tools and procedures to tap the 
construct; this topic will be covered under a separate heading and 
will include measures of ethnicity. 

Adaptation Processes 

Adjustment to another cultural lifestyle brings about significant 
changes for some and little change for others. Change is not an 
automatic outcome of resettlement, migration, relocation, and con- 
tact (Berry, Trimble, & Olmedo, 1986). For, indeed, individual varia- 
tions exist in response to the change process. Kagan and Cohen 
(1990) found among a sample of international students that cul- 
tural adjustment was affected by employment, language use, friend- 
ship patterns, work values, and decision-making patterns; cultural 
assimilation and transmutation accounted for a majority of the stu- 
dents' adjustment problems. Study participants, moreover, reported 
varying levels of problems for each adjustment category Such find- 
ings are consistent with the acculturation literature. 

In studying acculturation, several researchers prefer to con- 
sider the differential nature of the construct. Hertz (1984) argues 
that migration and the acculturative process are accompanied by 
a crisis orientation wherein emotional strain and trauma expose 
the individual and the family to stress. National and ethnic loyal- 
ties, too, often are tested. Alva (1985) found that low-acculturated 
individuals tended to express less satisfaction with the politics of 
the host country and more loyalty to their own country; there- 
fore, her findings support the notion that socialization in another 
country does not follow a linear progression. Similarly, Feldman 
and Rosenthal (1990) found that the process of acculturation was 
gradual for high school Chinese youth and that it varied as a func- 
tion of setting and age-related expectations. 

In addition to the broad-based effects created by the 
acculturative process, some researchers have examined more spe- 
cific variables and have found that simply endorsing a work ethic 

or learning the host language does not necessarily imply that full 
assimilation will occur (see Young & Gardner, 1990). Even an op- 
pressive, discriminating culture is not strong enough to advance 
sociocultural change. 

Acculturative Stress 

Acculturative stress can be defined as experiencing psychic diffi- 
culties in relation to changes in one's cultural surroundings, ei- 
ther by an individuars entry into a new culture or by the 
encroachment of a new culture on an already existing culture. It 
can be most easily understood as a negative reaction to encroach- 
ing Westernization or modernization. These psychic difficulties 
can range from the mildly pathological to problems of mental 
health and psychosomatic symptoms (Berry, J.W., 1980). 

Berry first used the term acculturative stress in 1971. But 
acculturative stress, as a concept, had been alluded to as early as 
1928 in anthropological and sociological journals. For instance, 
Robert E. Park (1928) spoke about how human migration created 
marginality Gillin, as quoted by Keesling in 1953, alluded to the 
problems experienced during acculturation: 'Tor every society 
undergoing acculturation there would seem to be an inevitable 
period characterized by some confusion and lack of stability in 
behavior.... Rapidly recurring, capricious and disorderly alter- 
ations of conditions will result in random behavior and cultural 
disorganization, apathy or withdrawal.'' Other authors have also 
indicated interest in research that might arise from a study of the 
effects of acculturation and the stresses arising from it, for both 
processes occurring zvithin acculturating individuals and interper- 
sonal processes occurring between acculturating individuals 
(Chance, 1965). Berry can be given credit for defining and crystal- 
lizing a concept that had been on the edge of research for nearly 
40 years. He can also be credited with doing most of the research 
on acculturative stress itself. 

In Berry's concept of acculturative stress, ecology, or a person's 
surroundings, plays a large part in the degree of acculturative 
stress suffered. Berry claimed that ecology "has a role in shaping 
human behavior ... it asserts the ecological limitation of behav- 
ioral development, the ecological source of the probability of be- 


havior or the behavioral adaptation to ecological pressures'' (Berry, 
1971). Ecology is instrumental, in individuals or in groups, for 
the development of our enculturation (how one's society is shaped). 
Acculturation and acculturative stress are reactions of old culture 
against new culture. The ecological setting helped shape that cul- 
ture. Hence, an ecological perspective is important in understand- 
ing acculturative stress. 

Berry has done the majority of the research on acculturative 
stress, although he is not the only researcher to pursue an under- 
standing of the subject. One possible reason for the limited amount 
of research on acculturative stress as a phenomenon is that it seems 
to be somewhat psychological in nature, given that people can 
suffer from varying degrees of psychic distress as a result of ac- 
culturation. Acculturative stress cannot be said to work in exact 
formulas, because people react to different situations in different 
ways. However, some generalizations can be made about the func- 
tioning of acculturative stress. The basic assumption is that the 
more similar an individual's or group's behavior is to the encroach- 
ing culture's behavior expectations, the less adjustment is neces- 
sary on their part, and, consequently, the less acculturative stress 
suffered. Furthermore, the assumption is made that individuals 
within a society who are more independent of the society (field 
independence) will be more independent of the incongruity and 
conflict arising from acculturation (Berry, 1975). 

A number of researchers have targeted specific ethnic groups 
in an effort to identify stress correlates and the acculturative pro- 
cess. Dressier and Bernal (1982) found that among Puerto Rican 
immigrants the length of residence mitigated against acculturative 
stress when psychosocial resources were available; if the resources 
were not available, migrants were likely to experience poor health 
and behavioral problems. But the availability of social support 
for immigrants may not be a sufficient deterrent against experi- 
encing stress-related problems. In a sample of Mexican immigrant 
women, Salgado de Snyder (1987) found a high degree of depres- 
sive symptomatology regardless of the social support available. 
More stress was experienced by respondents with a high degree 
of ethnic loyalty, lower levels of self-esteem, and lower personal 
satisfaction with life's circumstances. 

A series of studies report the relationship between accultura- 
tion and stress among immigrant Asians residing in the United 
States. These studies verified the general assumption that those 
least acculturated experienced the highest number of negative life 
events and that the most acculturated experienced the least 
amount of stress and the greatest amount of self-esteem (see Yu & 
Harburg, 1980; Yu, 1984; Padilla et al, 1985). 

Family Functioning 

Families subjected to the acculturative process seemingly are af- 
fected in ways similar to individuals. The family as a unit can be 
affected, and individual members experience varying levels of 
difficulty in adapting and adjusting to a new culture. In general, 
the younger the family member the less difficulty he or she will 
experience. However, level of adjustment at the group and indi- 
vidual level can vary as a function of the importance placed on 
familism. Sabogal, Marin, Otero-Sabogal, and Marin (1987) found 
three basic familism dimensions among samples of Hispanics and 
White non-Hispanics that include familial obligations, perceived 
support from the family, and family as referents. As the 
acculturative process unfolds, the "high level of perceived family 
support'' dimension remains stable and constant, whereas the 
other two dimensions are likely to diminish in importance. 
Rueschenberg and Buriel (1989) also found that as families of 
Mexicans proceed toward assimilation, they become increasingly 
involved with social systems outside the family; the internal fam- 
ily system of support, however, tends to remain stable. 

Acculturation and Counseling Services 

One's acculturative status can influence a decision to seek coun- 
seling, especially during times of confusion and duress. For ex- 
ample, the willingness of Asian Americans to see a counselor was 
related to acculturative status, gender, and ethnic identification; 
low- or medium-acculturated students were more likely than high- 
acculturated students to see a counselor (Gim, Atkinson, & 
Whitely, 1990). Bemak (1989) suggests that culturally different cli- 
ents, especially Southeast Asian refugees, experiencing accultura- 
tion will progress through developmental phases involving 


security and safety, integration of self and family, and future iden- 
tity. All three processes should be addressed in counseling and 
therapeutic settings. 

Drug and Alcohol Use and Acculturation 

Several studies have been conducted to identify the influence of 
acculturative status on alcohol and other drug use among immi- 
grant populations. In general, studies tend to suggest that immi- 
grant populations are likely to engage in any one of the following 
alcohol use patterns: (1) assimilating the ''core'' drinking patterns 
characterized by White middle-class America; (2) developing a 
new alcohol use style that follows a "melting pot" view; or (3) 
maintaining an alcohol use style reflective of use in the home coun- 
try. In a study of over 1,400 Italian- American adults, Blane (1977) 
found that a drinking style emerged that was a blend of their home 
country patterns and those characteristic of a typical American. 
He states that "the phenomenon of relatively high frequent drink- 
ing rates appears to be an instance where a recently learned cul- 
tural element combines with a declining but still powerful old 
cultural form to result in the acceleration of a drinking pattern to 
levels greater than usual in the host country" (p. 1,339). As such, 
ethnic-specific alcohol use patterns may not be encompassed by 
conventional acculturation models. Extensive variation exists in 
the motives and patterns for alcohol and other drug use and ac- 
culturation, for all of its complexity, contributes in some differen- 
tial way to the outcome. 

Early studies of the relationship between alcohol use and ac- 
culturation focused on Jewish, Irish, and Italian populations. In 
the past decade, though, emphasis has shifted to examining the 
relationship primarily among Mexican Americans. Neff, Hoppe, 
and Perea (1987) found in a sample of 164 non-Hispanic Whites 
and 149 Mexican- American male drinkers that the quantity of al- 
cohol use was significantly higher among less assimilated Mexi- 
can Americans; "escape" drinking motives were higher for this 
group, too. The authors point out, as did Blane, that a straightfor- 
ward acculturation model is not sufficient to explain alcohol use 
among ethnic minorities. In lieu of the conventional approach, 
the authors recommend use of a cultural /marginality stress model. 


A review of the literature on modal patterns of alcohol use 
among Mexican Americans emphasizes the importance and sig- 
nificance of using an acculturation stress model to explain the 
etiology of the issue. How Hispanics define their ethnic identity 
should be included in understanding the acculturative factors 
involved in alcohol use. Caetano (1986) identified four Hispanic 
definitions: family origin; national group; country most ancestors 
came from; and birthplace. The respondents who used the least 
amount of alcohol defined themselves by their birthplace. Thus, 
a definite link was established between ethnic identity and accul- 
turation. Self-efficacy and family interaction patterns are possible 
contributing variables in studies of the effects of acculturation on 
drug use. Sabogal et al. (1989) found that acculturation was in- 
versely related to self-efficacy but related to nicotine addiction 
among Hispanics. As Hispanics become assimilated, cigarette 
addiction tends to mirror that of the dominant culture. Although 
self-efficacy may not contribute to understanding the relationship 
between acculturation and drug use by Hispanics, Bonnheim and 
Korman (1985) point out that families that are perceived as con- 
fused, negativistic, inconsistent, and in internal conflict tend to 
promote inhalant abuse; furthermore, familial acculturative pat- 
terns may be disruptive to a child and thus contribute to the over- 
all social efficacy of the family. Substance abuse may be one way 
for youths to deal with the conflicts that can be brought about by 
acculturation stresses and difficulties. 

Acculturation and Successful Adaptation 

Much of the acculturation literature tends to emphasize negative 
and unhealthy adjustment outcomes. There are a few researchers, 
however, who point out that acculturating groups can adapt suc- 
cessfully to new environments. We must remember that individu- 
als in plural societies develop attitudes about the society as a whole 
and how they wish to relate to individuals and groups (Berry, 
Trimble, & Olmedo, 1986). Acculturation attitudes largely deter- 
mine orientations and perceptions of one's own group and rela- 
tionships with other groups. According to Berry (1983), individuals 


can hold acculturation toward any of the following dimensions: 
assimilation, integration, separation, and marginalization. 

A review of the definitions of adjustment, adaptation, accul- 
turation, assimilation, and effectiveness, especially as they relate 
to intercultural effectiveness, generates several skills and traits 
that contribute to successful or effective adaptation, including an 
ability to communicate, an ability to establish and maintain rela- 
tionships, an orientation toward knowledge, linguistic ability, flex- 
ibility, a realistic view of the target culture, and cultural empathy 
(Hannigan, 1990). Literature review findings also suggest that such 
factors as dependent anxiety, perfectionism, ethnocentrism, rigid- 
ity, narrow-mindedness, and self-centered role behaviors can con- 
tribute to negative adjustments. Mendenhall and Oddou (1985) 
also point out that expatriate acculturation is a multidimensional 
process that includes self-oriented, others-oriented, perceptual, 
and cultural-toughness dimensions. Such factors as relationship 
development, willingness to communicate, and stress reduction 
activities can promote effective adaptation. Groups and individu- 
als can anticipate the effects of the acculturative process if culture 
contact is imminent. Training and orientation programs should 
be developed to assist individuals in making effective and posi- 
tive adjustments to new settings. 

Measurement of Acculturation 
and Ethnic identification 

Measures of ethnic identification and acculturation are receiving 
considerable interest in cross-cultural and multicultural studies. 
Much of the research suggests that individuals experiencing a vari- 
ety of adaptation and adjustment problems are also experiencing 
difficulty with their ethnic identity. Pressures and stressors created 
by contact and interactions with another culture (or cultures) also 
appear to be intruding on adaptation. To assess culture-contact 
experiences many researchers rely on measures of acculturation. 
Moreover, some researchers suggest that ethnic identification and 


acculturation may be responsible for adjustment problems but 
offer no data to support their views. The literature on measuring 
acculturation yields three overriding findings: (1) measures of the 
acculturation process vary considerably in length, theme, and scal- 
ing properties; (2) typically acculturation, as a construct and a 
concept, is equated with adaptation and therefore is used in very 
broad terms; and (3) reliance on embedding ethnic-identity mea- 
sures in acculturation measures is increasing. 

Acculturation Measures 

A close inspection of a number of acculturation measures indi- 
cate that many researchers view the construct as a unidimensional, 
linear phenomenon; hence, the measure yields a single score that 
identifies one's position on a nativist-traditional/assimilationist 
dimension. Other researchers emphasize that the acculturative 
process is multifaceted, bidirectional, and situational; measures 
following this theoretical perspective produce composite rather 
than single scores and measures. 

More often than not, the item content and number of items 
used to measure acculturation are similar, regardless of the theo- 
retical position taken by the researcher. Some measures contain 
as few as 3 items while a few others contain as many as 43 items; 
this discrepancy, in itself, can raise serious validity questions. Ac- 
culturation scale content typically includes one theme or a com- 
bination of three themes: natality, behavioral predisposition, and 
subjective preferences. All three themes can contain items that 
often are f oimd in measures of ethnic identification. Some research- 
ers have identified an ethnic identity factor embedded in their 
acculturation scales — we continue to find researchers using items 
in acculturation scales that also are found in ethnic identity scales. 
The preceding measurement issues are but a few of the emerging 
concerns about measuring the two constructs. Brief reviews of 
the literature on the topic follow. 

Psychological Acculturation 

In studying and assessing the acculturative process, cross-cultural 
psychologists prefer to place an emphasis on an individuars ex- 


periences. As a consequence, psychologists usually refer to the 
acculturative process as psychological acculturation. The bulk of 
the studies flowing from this orientation use measures and scales 
that attempt to isolate an individual's cultural orientation on a 
bipolar linear continuum. Through an analysis of responses, an 
individual is placed somewhere between a nativist or traditional- 
ist pole and a fully acculturated position. Subsequent compari- 
sons are typically performed on some other social and 
psychological variable where the investigator presumes that some 
sort of collinearity exists between acculturative status and a de- 
pendent variable. 

An example of a cultural contact measure is the eight-variable 
scale developed by Berry et al. (1985) for use in Central Africa. 
The scale tapped such contact-specific information as number of 
local languages spoken, property and material ownership, em- 
ployment status, clothing preferences, and religiosity. Responses 
to the scale were collapsed to a single score from which an infer- 
ence about the individuars acculturative status was derived. 
Ranch, Bowler, and Schwarzer (1987) developed a three-item cul- 
ture contact scale to assess acculturation effects on the self-esteem 
of three American ethnic minority groups. Although Ranch et al. 
found significant differences for self-esteem between the groups, 
they also reported an extremely low correlation between contact 
and self-esteem (.12), suggesting that contact may not influence 
one's sense of self. 

Triandis, Kashima, Shimada, and Villareal (1986) used a par- 
ticipatory measure of acculturation to assess the existence of cul- 
tural differences among a sample of Hispanic and non-Hispanic 
Navy recruits. Triandis et al. used a four-item index, including 
items that assessed length of residence in the United States, me- 
dia acculturation (e.g., preference for Spanish- rather than English- 
language television, radio, and movies), number of non-Hispanic 
coworkers, and number of non-Hispanic friends and romantic 
partners. A single acculturation score was created by summing 
the four indices. The study's Hispanic participants were grouped 
according to the single culture index into low-, medium-, and high- 
acculturation categories. The authors conclude that "one can use 
indexes of acculturation to establish the existence of cultural dif- 

ferences'' (p. 67). Pumariega (1986) also used a 15-item contact 
and participation acculturation scale to explore the relationship 
between socioeconomic status (SES) and eating attitudes among 
Hispanic adolescent girls. His results showed that the more one 
moves toward a contact culture, the greater is one's vulnerability 
for developing an eating disorder Pumariega also found that SES 
is not correlated with acculturative status, suggesting that one's 
occupational and educational status may not be useful in under- 
standing the influences of cultural contact and participation. His- 
panic populations, more than any other ethnic group, appear to 
be the focus of interest among acculturation researchers. 

In addition to the scales described above, measures have been 
developed for Mexican- American children, adolescents, and 
adults (see Franco, 1983; Martinez, Norman, & Delaney, 1984; 
Cuellar et al, 1980). Mendoza (1984) developed an elaborate Cul- 
tural Life Style Inventory to assess acculturative status among 
Mexican- American adolescents and adults. The inventory taps five 
factor dimensions: intrafamily language usage; extrafamily language 
usage; social affiliations and activities; cultural identification and 
pride; and cultural resistance, incorporation, and shift. The scale con- 
tents capture much of the same information developed in other 

Richman, Gaviria, Flaherty, Birz, and Wintrob (1987) devel- 
oped a 21-item scale that tapped five dimensions — language 
(6 items); customs (4 items); ethnic identity (1 item); sociability 
(4 items); and discrimination (6 items) — for use among 
Huahuapuquien Indian migrants in Lima, Peru. The multifaceted 
scale was used to assess the acculturative process in the Peruvian 
social-structural context. Of importance is the researchers' attempt 
to assess the effects of discrimination on adaptation. 

Suinn, Rickard-Figueroa, Lew, and Vigil (1987) developed an 
acculturation scale for assessing Asian acculturation; the scale 
contents were modeled after several scales developed for use with 
Hispanics. Researchers have been able to validate the Suinn-Lew 
scale in applied settings. 

Two studies are reported in the literature that draw attention 
to the development of more generalized acculturation scales that 
transcend ethnic-specific measures. Oakland and Shermis (1989) 


analyzed the factor structure of the Sociocultural Scales of the 
System of Multicultural Pluralistic Assessment. Their analysis 
supported the existence of four subscales, one of which was urban 
acculturation. The latter subscale is a measurement domain that 
heretofore has received little attention in the acculturation field. 
An American International Relations Scale developed by 
Sodowsky and Plake (1991) represents the current state of the art 
in measuring acculturation. The University of Nebraska-based 
researchers maintain that seven moderator variables influence the 
acculturation process: (1) generational status; (2) education and 
income; (3) age; (4) years of residence in the United States; 
(5) ethnic density of neighborhood; (6) country of birth; and 
(7) job skills, religion, kinship structures, and purposes of 
immigration. Their 34-item scale has been expanded to include 
43 items and is referred to as the Majority-Minority Relations 
Survey (Sodowsky, Lai, & Plake, 1991). Results from their early 
work with the scale are favorable and suggest that a valid, 
generalizable, and reliable scale has been developed to assess 
acculturation attitudes of Asian Americans and Hispanics^ 

Promoting a Multilevel Approach 

Most measures of contact-participation acculturation generate a 
single average or cumulative score. And, to some extent, the single- 
unit scales have proven to be valid and hence useful largely be- 
cause certain researchers have been able to make distinctions 
between and among cultural groups and predict outcomes such 
as mental health, alcohol and other drug use, and eating disor- 
ders. The use of a single score is presumptive since it considers all 
who fall at or about the unit to be relatively homogeneous. That 
may not be the case. 

Berry, Trimble, and Olmedo (1986) argue that ''not every per- 
son in the acculturating group will necessarily enter into the ac- 
culturation process in the same way or to the same degree'' (p. 
296). Not every individual from his or her culture will respond to 
the elements presented by the dominant culture. Individual varia- 
tions in the adaptation, adjustment, and internalization of another 
culture's folkways and mores can be mediated by resistance, fear, 
anxiety, allegiance to one's own culture, and the level of perceived 


acceptance in one's own and the dominant culture. Moreover, an 
individual may make specific accommodations in certain settings 
and situations and in others cling to and maintain behaviors that 
are traditional and conventional in the individual's own culture. 
As a consequence, one's adjustments and accommodations, com- 
bined with the rejection of acceptable behavioral patterns, gener- 
ates a good deal of variation in acculturative styles. 

Questions concerning the unidirectional issue and individual 
variations in acculturative styles, combined with the use of single- 
unit measures of acculturation, are beginning to emerge in the 
literature. J.W. Berry (1980) takes the position, for example, that 
acculturation can be viewed as a multilinear phenomenon rather 
than a single dimension. Using a Likert-scale format. Berry de- 
veloped a series of items that ultimately generate four separate 
cumulative scores, thus providing a more inclusive understand- 
ing of one's attitude toward acculturation and one's position in 
the process. 

Padilla (1980) proposed an acculturation framework that em- 
braces elements of the contact-participation dimension and one's 
perceived loyalty to one's own culture. Padilla uses 11 dimen- 
sions (e.g., language preferences, name preference for children) to 
measure loyalty and 15 dimensions to identify cultural aware- 
ness. Padilla concluded that "cultural awareness is the more general 
component" and "ethnic loyalty is the more tenuous" (1980, p. 65). 

Padilla's two-factor model formed the basis of a multifacto- 
rial acculturation model developed by Richman et al. (1987). These 
authors agree that "the process of acculturation in Peru is com- 
posed of a more complex set of sub-components, each of which 
may have very different social meanings, covary separately for 
different social status groups, and have differing consequences 
for psychological distress or well being" (p. 842). As a consequence 
of their observations of the various Peruvian cultural groups, 
Richman et al. created an overall acculturation scale containing 
four subscales: (1) language use, ethnic customs, encompassing 
music, food preferences, and dress; (2) ethnic identification; 
(3) sociability preferences; and (4) perceived discrimination. The 
scale generates six distinct scores. Scale results were useful in 
identifying variations across age and generation levels and 


showing that changes among certain individuals are bidirectional 
and characterized by inequality. 

A multivariate perspective, as recommended by Berry (1983), 
would indeed provide useful information about the source of in- 
fluences as the acculturation process is experienced. However, 
such knowledge would provide us with information only about a 
person's style, preferences, behavior, and orientation. Ethno- 
graphic- and case-directed interviews of acculturating individu- 
als generate observations that point to the fact that the context in 
which acculturation occurs and the situations one encounters of- 
ten dictate unique behaviors that in themselves are situation spe- 
cific. Given the observations, it would seem reasonable to propose 
that more attention should be given to the situations and the con- 
texts in which acculturation occurs. 

As individuals experience a new culture, they will undoubt- 
edly encounter new and varied situations. For those interested in 
the acculturative process, it makes sense to focus on the individu- 
als' perceptions of any situation they encounter, the behavior they 
choose to evoke, and their cognitive appraisals of the effect the 
situation holds for them. Interactionists, or those who study the 
person-situation interaction, recognize that individuals often will 
select the situations they encounter. Many situations may not be 
selected simply because they do not know what to do or how to 
react. Individual variations in situation selection may be directly 
related to acculturative status. Hence, the selection of situations 
may shed some insights on the types of situations selected and con- 
sequently further our understanding of the acculturative process. 

An interest in expanding the measurement of acculturation is 
emerging that will include domains other than those involving 
participation. The work of Berry (1983), Padilla (1980), and 
Richman et al. (1987) is a testament to that interest. The recent 
theoretical work of Getting and Beauvais (1991) and Pumariega 
(1986) points to some other potentially significant developments. 

To further our understanding of the measurement of the 
acculturative process, Trimble (1988b) proposes a model that in- 
cludes many of the elements discussed elsewhere in this chapter. 
Along with these elements, it includes a series of discrete elements 
that emphasize the situations that individuals encounter — par- 


ticularly those that involve forms of interethnic discrimination 
and racism. The model is built on a certain set of assumptions 
emerging from causal modeling theory and what we know cur- 
rently about measuring acculturation. The model is cast in a path 
analytic framework because we are interested in identifying a set 
of specific relations among a set of variables that could predict 
one's acculturative status. The use of path analysis as a math- 
ematical modeling technique assumes that a set of variables can 
be temporally ordered, are asymmetrically related, and are mea- 
sured on at least an interval scale, and that the relations among 
the variables are linear and additive (Duncan, 1975). The hypo- 
thetical multilinear model is illustrated in Figure 1. The straight 
lines linking the variables represent causal assertions. The math- 
ematical notations represent path coefficients associated with each 
line and are hypothetical estimates of the magnitude of the effect 
an explanatory variable has on the variable to which it is point- 
ing. These relations are viewed as independent of other explana- 
tory variables. A single path or a combination of paths can lead to 
a prediction of the dependent variable, in this case the list of four 
acculturation alternatives at the extreme end of the model. In its 
present form the model is tentative, hypothetical, and therefore 
subject to discussion, validation, and refinement. 

Ethnic Identification Measures 

As pointed out in the previous section, several researchers found 
it necessary to include measures of ethnic identity in their respec- 
tive acculturation scales. Such subscales range from one item 
(Richman et al., 1987) to several items. Yet the literature contains 
several articles that address the need for developing unique mea- 
sures of ethnic identification. 

Although this appears to be a straightforward recommenda- 
tion. Smith (1980) reminds us that ''ethnicity is not a simple all or 
nothing proposition. Researchers have long recognized that a 
person's level or intensity of identification with a particular 
ethnicity can vary from a weak-nominal association to a strong- 
committed association" (p. 79). 




H C/5 

< 2 

CsS Ci 

3 F 


















O ml 





















«•■=■« .2 
= 2 £ tl 
£ ta tb y^ 


.9- g = 

'5 T3 > 


.y -D ;> 

(2 -^ < 












Ethnicity can be measured by framing questions that focus on 
one or any combination of three domains: (1) the natal, where 
inquiries are made about one's place of birth and the birthplaces 
of parents, grandparents, and relatives; (2) the behavioral, where 
information is obtained on language usages, affiliative patterns 
with friends and acquaintances, media use, and participation in 
ethnic specific activities; and (3) the subjective, where the per- 
sons are asked to indicate, for instance, what ethnicity they con- 
sider themselves to be, where relatives and ancestors came from, 
and their ethnic preferences (Smith, 1980). 

Most measures of ethnic identity provide little or no opportu- 
nity to assess the variable and fluctuating nature of the concept. 
One's identification with an ethnic group is not static or immu- 
table. It can and does change over time as a function of varied 
circumstances and social contexts. There are situations in which 
people tacitly identify with significant others that are associated 
with the contexts, and then there are other settings where an indi- 
vidual may act quite differently because the ethnic makeup may 
be quite different. When individuals vary their behavior, they may 
vary their level of group association from a weak-nominal, al- 
most passive level to a strong-committed, active one. Hence, it 
maybe that one's intensity of ethnic identification fluctuates with 
a particular social context or situation (Trimble, in press). 

Perhaps the most widely used measurement procedure in- 
volves the use of a list of ethnic groups on which respondents are 
merely asked to place a check mark next to the group with whom 
they most identify. Use of the procedure hardly qualifies as a scale; 
but more than that, it is replete with erroneous assumptions. First, 
it forces a person with a multi-ethnic background to choose one 
group. Second, selection of a group does not connote the inten- 
sity of identification. Third, the procedure often generates numer- 
ous ethnic categories to the point where stratified and subgroup 
analyses may not be possible. 

Ethnic identity scales assume that one can readily identify with 
a group or even prefers to do so. Hence, use of an item that asks 
one to check or list the group with which one identifies is a good 
starting point — most multi-item identity scales start off with this 
question. Phinney's (1992) Multigroup Ethnic Identity Measure 


actually asks respondents to indicate their ethnic affiliation twice 
in the 23-item scale. Her scale, though, assesses self-identification 
through two dimensions: ethnic identity and other group orien- 
tation. The ethnic identity dimension also taps affirmation, be- 
longing, ethnic identity achievement, ethnic behaviors, and natural 
parents ethnicity; the latter items are not scored and, therefore, 
are used only as background information. 

According to Weinreich (1986), one's ethnic identity is "situated 
in a specific social context [and] is defined as that part of the totality 
of one's self construal'' (p. 2). Weinreich has been exploring the vari- 
able nature of situated identities through the use of Identity Struc- 
ture Analysis (ISA), a method that permits the operationalization of 
''a person's conflicted identification with another, so that the rela- 
tive magnitudes of self's various identification conflicts may be esti- 
mated" (p. 300). As a result of his extensive work on the subject, 
Weinreich strongly suggests that "(1) individuals can indeed be some- 
what different beings according to which ethnic context they cue 
into; (2) when cued into the one ethnicity, the values characteristics 
of the alternative ethnicity may be emphasized; and (3) in ... linguis- 
tic contexts, the identity structures of individuals are essentially the 
same kind of complex amalgamation of identification elements across 
... ethnicities" (pp. 305-306). 

ISA involves the use of a complex array of measurement items 
and therefore could be time consuming. It, too, necessitates that 
one deals with the specific situations one encounters in order to 
determine one's level of identification. Nonetheless, the argimients 
for using an abridged version of ISA in the field are compelling. 

Researchers have resorted to the use of less obtrusive mea- 
sures in an effort to tap ethnic identity. Most researchers are acutely 
aware of the seminal doll selection technique developed by Clark 
and Clark (1947). Basically, Black (African-American) children 
showed a preference for White dolls, suggesting that the youth 
identified more with Whites than with Blacks. Clark and Clark's 
work, especially the methodology and conclusions, has been ques- 
tioned by other researchers (see Katz & Zalk, 1974; Stephan & 
Rosenfeld, 1978); however, use of the technique persists. For ex- 
ample, although aware of the flaws, Vaughan (1986) demonstrated 


that Maori children in New Zealand select pictures and dolls that 
represent their image of their group within the dominant culture. 
The article also contains a review of the literature on using pic- 
ture and doll tests. 

Fine and Bowers (1984) replicated the design and procedure 
of the Clark and Clark study. Their findings revealed that young 
African- American females were more likely to identify with Black 
dolls than were young African- American males. The socioeco- 
nomic and political climate, the authors suggest, may contribute 
to their findings; such factors must be considered when results 
from doll selection techniques are used. 

Helms (1989) developed and tested both a Black and a White 
racial identity scale; both scales use prejudice and discrimination 
type items to assess the identity construct. Moreover, Black and 
White identity measures are based on historical and current in- 
terracial attitudes and behavior toward one another. Hence, iden- 
tity is defined as "attempts to explain the various ways in which 
Blacks can identify (or not identify) with other Blacks and /or adopt 
or abandon identities resulting from racial victimization; White 
racial identity theories attempt to explain the various ways in 
which Whites can identify Whites and /or evolve or avoid evolv- 
ing a nonoperative White identity'' (p. 5). 

Getting and Beauvais (1991) developed a series of accultura- 
tion measures to test their bidimensional model of acculturation. 
These two psychologists agree that one's identity with a cultural 
group can be directional where one can be placed on a linear con- 
tinuum ranging from high to low. Using a sample of nearly 2,400 
American-Indian youths, they found that their subjects could be 
located in a quadrant on a multidimensional plane. Some showed 
a low White-high Indian preference, some showed a low Indian- 
high White preference, and still others showed a low Indian-low 
White preference. The finding suggests that the third group of 
subjects preferred not to identify and hence participate in either 
the Indian or the White culture. Beauvais and Getting suggest 
that yet another plane can be created by adding another cultural 
group, such as Mexican, to substantiate further that one chooses 
a cultural group on the basis of the stake one has in that culture. 


Using the American-Indian cultural identification items de- 
veloped by Getting and Beauvais (1991), we tested the tripartite 
model of ethnic self-identification (Trimble, 1991; Bates, 1994). The 
items were treated with a restrictive, principal components factor 
analysis, which produced a single factor or latent variable. Figure 
2 shows the seven measurement domains and their correspond- 
ing factor loadings. Results suggest that items tapping natal, be- 
havioral, and subjective identification domains form a single latent 
construct, which can subsequently be used to measure Indian eth- 
nic self-identification. Because of procedural and methodological 
considerations, the items were not used with tribal-specific 
samples; further work in this area should be conducted to learn 
more about intertribal differences and similarities. 

The scale was subsequently used to predict alcohol involve- 
ment among American-Indian youths. A sample of 696 Indian ado- 
lescents from five rural, reservation, and urban environments in 
the southern plains and southwestern region of the United States 
completed a 52-page survey developed by the Tri-Ethnic Center 
for Prevention Research at Colorado State University. Youths rep- 
resented high school dropouts, academically at risk nondropouts, 
and nondropouts. 

Four independent measures were used in the analysis: gender; 
peer alcohol association; family alcohol association; and the ethnic 
self-identification scale (see Bates, 1994). Each measure was con- 
firmed using a variant procedure of confirmatory factor analysis. 
Structural latent variable analysis was used to generate solutions 
for the measurement models and to validate the structural equa- 
tion model that appears in Figure 3. The Comparative Fit Index 
(CFI) of .901 was adequate and the chi-square of 255.10 (df=58, 
p<.001) was significant, revealing a nonexact fit of the model; the 
large sample size of 696 may account for the latter result. Hence, 
the relationship between each independent variable and alcohol 
involvement was significantly different from no relationship. 

The sole predictor of membership in an academic status group 
was alcohol involvement; thus, we concluded that as alcohol in- 
volvement of an Indian adolescent increases, so does the likeli- 
hood of having academic problems or dropping out of school. 
Additionally, American-Indian identity and gender status were 










TJ ^ 

^ -r 


t: c 

gj c 

Si c 


U5 2 








O -Jj 

C c 

c -2 





^ s 











P q3 



^ ro 






•o = 

S a. 




ri^ ^ 














° 1 















^ CT) 
O O) 


found to be unrelated to alcohol involvement. Consequently, the 
finding suggests that degree of ethnic self-identify is not a predic- 
tor of alcohol involvement for Indian adolescents. To the contrary, 
peer alcohol associations and, to a much lesser extent, family al- 
cohol associations predict alcohol involvement. 

The finding that ethnic self-identity is not associated with al- 
cohol involvement is by no means conclusive. Peer associations 
do influence level of alcohol involvement — one could assume that 
most of the sample's peers were American Indian, hence some 
cultural or ethnic factor may be at work in mediating alcohol use. 
Our findings warrant further study with other samples of Ameri- 
can-Indian youth and other ethnic groups, especially as they re- 
late to drug involvement. 

As previously indicated, it appears that measuring accultura- 
tion and ethnic identity is no small matter. Undoubtedly there is 
considerable overlap between the two constructs. One doubts in- 
deed that measuring one construct involves the use of elements 
from the other. The whole topic is worthy of considerable debate 
and research. 

Sumnnary and Conclusions 

Change is an inevitable human and environmental condition. At 
sociocultural and individual levels change influences all aspects 
of humankind. For most humans the change process is orderly, 
cyclical, and predictable. Societal elements can intrude and inter- 
fere with the process, disrupting the patterns and causing disar- 
ray for the populace. From a sociocultural and psychological level 
of analysis, isolating the cause-and-effect relationships between 
and among the intruding and disrupting elements is an enormous 
challenge. The challenge is even more formidable when cultural 
and ethnic groups are forced to immigrate to countries where the 
cultural orientation is completely different; the change can be dis- 
ruptive enough to affect all aspects of the immigrating groups. 

Acculturation is a form of sociocultural change. Originally 
identified and conceptualized by anthropologists, the concept now 
is included in the research agenda of psychologists, psychiatrists, 
sociologists, social workers, and educators. As a consequence of 


the increased interest in the topic, acculturation's meaning and 
its application are changing, as are techniques and procedures for 
measuring the concept. For the most part, researchers have at- 
tempted to attribute the results of sociocultural change to the 
acculturative process. The attribution process has been extended 
to blaming negative adaptation and adjustment on acculturation 
as though it has a direct effect on outcomes. Such attributions, 
however well meaning, have served to confound the research 
process and muddle the field of inquiry. Yet there is no doubt that 
when two or more intact cultural groups come into direct contact, 
conflict is one inevitable outcome. How individuals and groups 
deal with the contact and conflict continues to be an important 
and significant research question. Practical and applied aspects 
of the research are also important considerations. 

Ethnic identification is related to acculturation. Several re- 
searchers, for example, have included measures of ethnic identity 
in their acculturation scales. The acculturative process also height- 
ens an individuaFs sense of ethnic identity, perhaps more so than 
before cultural contact. The similarities do not end there, though, 
because the many variables that affect the acculturative process 
also seem to affect ethnic identity in much the same way. 

Numerous definitions abound for the concept of ethnic iden- 
tity. Some researchers prefer the term racial identity, and others, 
cultural identity. Several variations of the concept are discussed in 
this chapter, including hierarchical nesting and the "ethnic gloss'' 
notion. Like acculturation, measurement of identity is no small 
undertaking since it is fraught with many complications. 

The intent of this chapter is to provide a reasonably compre- 
hensive review of the many issues and topics mixed in with ac- 
culturation and ethnic identity. The chapter by no means is all 
encompassing and definitive, even though the topics and issues 
are salient. Moreover, the topics and issues for both concepts are 
similar and seemingly interchangeable. The concepts also are in- 
timately connected to culture, which by most criteria is really a 
metaphor. Could it be then that acculturation and ethnic identity 
also fall into the same category? And if that is the case then a 
good deal of time, resources, and energy have been devoted in a 
vain attempt to isolate an intangible. 



The author wishes to extend his gratitude to the staff at the Tri- 
Ethnic Center for Prevention Research at Colorado State Univer- 
sity for assistance and support in writing this chapter. Support 
also was provided through a grant from the National Institute on 
Drug Abuse (P50 DA 07074) and the National Institute on Alco- 
hol Abuse and Alcoholism (AA 08302). The author wants to espe- 
cially thank Scott Bates for his thoughtful and skillful analysis of 
the data presented in this chapter and Brad Plemons for his assis- 
tance in editing and compiling the references. 


Alva, S. (1985). The political acculturation of Mexican American adolescents. 
Hispanic Journal of Behavioral Sciences, 7(4), 345-364. 

Andujo, E. (1988). Ethnic identity of transethnically adopted Hispanic 
adolescents. Social Work, 33(6), 531-535. 

Baldwin, J., Duncan, J., & Bell, Y. (1987). Assessment of African self-consciousness 
among Black students from two college environments. Journal of Black 
Psychology, 13(2), 27-34. 

Bates, S. (1994) . A cross-validated model of the causes and effects of alcohol involvement 
on American Indian adolescents. Unpublished master's thesis. Western 
Washington University, Bellingham, WA. 

Bemak, F. (1989). Family therapy with immigrant families: Constructing a bridge 
between different world views. Journal of Strategic and Systemic Therapies, 8, 

Berry, J. (1980). Social and cultural change. In H.C. Triandis & R. Brislin (Eds.), 
Handbook of cross-cultural psychology. Vol 5: Social psychology. Boston: Allyn & 

Berry, J. (1985). Cultural psychology and ethnic psychology: A comparative 
analysis. In I. Lagunes & Y. Poortinga (Eds.), From a different perspective: Studies 
of behavior across cultures (pp. 3-15). Lisse, The Netherlands: Swets & Zeitlinger. 

Berry, J., Trimble, J.E., & Olmedo, E. (1986). Assessment of acculturation. In W. J. 
Lonner & J.W Berry (Eds.),f zeW methods in cross-cultural research (pp. 291-234). 
Newbury Park, CA: Sage. 

Berry, J.W. (1971). Ecological and cultural factors in spatial perceptual 
development. Canadian Journal of Behavioral Science, 3, 324-336. 


Berry, J.W. (1975). Ecology, cultural adaptation, and psychological differentiation: 
Traditional patterning and acculturative stress. In R. Brislin, S. Bochner, and 
W. Lonner, (Eds.), Cultural perspectives in learning (pp. 207-228). Beverly Hills, 
CA: Sage. 

Berry, J.W. (1980). Acculturation as varieties of adaptation. In A. Padilla (Ed.), 
Acculturation: Theory, models and some new findings. Boulder, CO: Westview. 

Berry, J.W. (1983). Acculturation: A comparative analysis of alternative forms. In 
R. Samuda & S.L. Woods (Eds.), Perspectives in immigrant and minority education 
(pp. 65-77). New York: Lanham. 

Berry, J. W., van de Koppel, J., Senechal, C, Annis, R.C., Bahucher, S., Cavalli- 
Sforza, L., & Witkin, H. (1985). On the edge of the forest: Cultural adaption and 
cognitive development in Central Africa. Lisse, The Netherlands: Swets & 

Blane, H.T (1977). Acculturation and drinking in an Italian American community. 

Journal of Studies on Alcohol, 38, 1324-1345. 

Blumer, H. (1969). Symbolic interactionism: Perspective and method. Englewood 
Cliffs, NJ: Prentice-Hall. 

Bonnheim, M., & Korman, M. (1985). Family and acculturation in Mexican- 
American inhalant users. Journal of Psychoactive Drugs, 27(1), 25-33. 

Broman, C, Neighbors, H., & Jackson, J. (1988). Racial group identification among 
Black adults. Social Forces, 67(1), 146-158. 

Bromley, M. (1988). Identity as a central adjustment issue for the Southeast Asian 
unaccompanied refugee minor. Child and Youth Care Quarterly, 17{2), 101-114. 

Caetano, R. (1986). Alternative definitions of Hispanics: Consequences in an 
alcohol survey. Hispanic Journal of Behavioral Sciences, 8, 331-344. 

Carter, R., & Helms, J. (1988). The relationship between racial identity and social 
class. Journal of Negro Education, 57(1), 22-30. 

Casino, E. (n.d.). Introduction to ethnicology: Ways of talking about ethnicity. 
Unpublished manuscript. Honolulu, HI. 

Chance, N.A. (1965). Acculturation, self-identification and personality 
adjustment. American Anthropologist, 67a, 372-393. 

Clark, K.B., & Clark, M. (1947). Racial identification and preferences in Negro 
children. In TM. Newcomb & E.L. Hartley (Eds.), Readings in social psychology. 
New York: Holt. 

Coelho, G.V., Hamburg, D.A., & Adams, J.E. (Eds.) (1974). Coping and adaptation. 
New York: Basic Books. 

Comas-Diaz, L., & Jacobsen, F. (1987). Ethnocultural identification in 
psychotherapy Psychiatry, 50(3), 232-241. 

Cuellar, I., Harris, L., & Jasso, R. (1980). An acculturation scale for Mexican- 
American normal and clinical populations. Hispanic Journal of Behavioral 
Sciences, 2, 199-217. 


Dohrenwend, B.S. (1973). Social status and stressful life events. Journal of 
Personality and Social Psychology, 28, 225-235. 

Dressier, W., & Bernal, H. (1982). Acculturation and stress in a low-income Puerto 
Rican community. Journal of Human Stress, 8{3), 32-38. 

Duncan, O.D. (1975). Introduction to structural equation models. New York: 
Academic Press. 

Dupuy, H.J., Engel, A., Devine, B.K., Scanlon, J., & Querec, L. (1970). Selected 
symptoms of psychological stress: United States. Washington, DC: U.S. 
Government Printing Office. 

Feagin, J.R. (1978). Racial and ethnic relations. Englewood Cliffs, NJ: Prentice- 

Feldman, S., & Rosenthal, D. (1990). The acculturation of autonomy expectations 
in Chinese high schoolers residing in two Western nations. International Journal 
of Psychology, 25(3), 259-281. 

Fine, M., & Bowers, C. (1984). Racial self-identification: The effects of social history 
and gender. Journal of Applied Social Psychology, 14(2), 136-146. 

Fisher, R. (1982). Social psychology: An applied approach. New York: St. Martin's 

Flaskerud, J. (1990). Matching client and therapist ethnicity, language, and gender: 
A review of the literature. Issues in Mental Health Nursing, 22(4), 321-336. 

Franco, J. (1983). An acculturation scale for Mexican- American children. Journal 
of General Psychology, 205(2), 175-181. 

Galan, F. (1988). Alcoholism prevention and Hispanic youth. Special Issue: 
Alcohol problems and minority youth. Journal of Drug Issues, 28(1), 48-58. 

Gim, R., Atkinson, D., & Whiteley, S. (1990). Asian-American acculturation, 
severity of concerns, and willingness to see a counselor. Journal of Counseling 
Psychology, 37(3), 281-285. 

Gonzalez, J. (1988). Dilemmas of the high-achieving Chicana: The double-bind 
factors in male/female relationships. Sex Roles, 28(7-8), 367-380. 

Grossman, B., Wirt, R., & Davids, A. (1985). Self-esteem, ethnic identity and 
behavioral adjustment among Anglo and Chicano adolescents in West Texas. 
Journal of Adolescence, 8{1), 57-68. 

Hannigan, T. (1990). Traits, attitudes, and skills that are related to intercultural 
effectiveness and their implications for cross-cultural training: A review of 
the literature. International Journal of Intercultural Relations, 24(1), 89-111. 

Helms, J. (1989). Considering some methodological issues in racial identity 
counseling research. The Counseling Psychologist, 27(2), 227-252. 

Helms, J. (Ed.). (1990). Black and White racial identity: Theory, research, and practice. 
Westport, CT: Greenwood. 


Herskovits, M. (1948). Mam and his works: The science of cultural anthropology. New 
York: Knopf. 

Hertz, D. (1984). Psychological and psychiatric aspects of remigration. Israel 
Journal of Psychiatry and Related Sciences, 21(1), 57-68. 

Judd, E. (1990). Intermarriage and the maintenance of religio-ethnic identity: A 
case study. Journal of Comparative Family Studies, 21(2), 251-268. 

Kagan, H., & Cohen, J. (1990). Cultural adjustment of international students. 
Psychological Science, 1(2), 133-137. 

Katz, D. (1974). Factors affecting social change: A socio-psychological 
interpretation. Journal of Social Issues, 30(3), 159-180. 

Katz, P., & Zalk, S. (1974). Doll preferences: An index of racial attitudes? Journal 
of Educational Psychology, 66, 663-668. 

Keesling, P.M. (1953). Cultural chance: Analysis and bibliography of anthropological 
sources to 1952. Stanford, CA: Stanford University Press. 

Kiev, A. (1972). Transcultural psychiatry. New York: Free Press. 

Klineberg, O., & Zavalloni. M. (1969). Tribalism and nationalism. The Hague: 

Kroeber, A. & Kluckhohn, C. (1952). Culture: A critical review of concepts and 
definition. Papers of the Peabody Museum of American Archeology and Ethnology, 

Leighton, A.H. (1949). Human relations in a changing world: Observations on the use 
of the social sciences. New York. 

LeVine, E., & Padilla, A. (1980). Crossing cultures in therapy: Pluralistic counseling 
for the Hispanic. Belmont, CA: Wadsworth. 

Looney, J. (1988). Ego development and Black identity. Journal of Black Psychology, 
15(1), 41-56. 

Manderscheid, R.W., Silbergeld, S., & Dager, E.Z. (1975). Alienation: A response 
to stress. Journal of Cybernetics, 5, 91-105. 

Marin, C, & Marin, B.V. (1982). Methodological fallacies when studying 
Hispanics. In L. Bickman (Ed.), Applied Social Psychology Annual (Vol. 3, pp. 
99-117). Beverly Hills, CA: Sage. 

Marris, P. (1974). Loss and change. New York: Pantheon. 

Martinez, R., Norman, R., & Delaney, H. (1984). A Children's Hispanic 
Background Scale. Hispanic Journal of Behavioral Sciences, 6(2), 103-112. 

Mays, V. (1986). Identity development of Black Americans: The role of history 
and the importance of ethnicity. American Journal of Psychotherapy, 40(4), 

McRoy, R., Zurcher, L., Lauderdale, M., & Anderson, R. (1984). The identity of 
transracial adoptees. Social Casework, 65(1), 34-39. 


Mead, M. (1963). Socialization and enculturation. Current Anthropology, 4, 

Mendenhall, M., & Oddou, G. (1985). The dimensions of expatriate acculturation: 
A review. Academy of Management Review, 10(1), 39-47. 

Mendoza, R. (1984). Acculturation and sociocultural variability. In J.L. Martinez 
& R. Mendoza (Eds.), Chicano Psychology (2nd ed., pp. 61-75). Orlando: 
Academic Press. 

Moore, D. (1990). Drinking, the construction of ethnic identity and social process 
in a Western Australian youth culture. British Journal of Addiction, 85(10), 

Moos, R.H. (Ed.). (1976). Human adaptation: coping with life crises. Lexington, MA: 

Morishima, J., Sue, S., Teng, L.N., Zane, N., & Cram, J. (1979). Handbook of Asian 
American/Pacific Islander mental health research. Rockville, MD: National Institute 
of Mental Health. 

Neff, J.A., Hoppe, S.K., and Perea, P. (1987). Acculturation and alcohol use: 
Drinking patterns and problems among Anglo and Mexican American male 
drinkers. Hispanic Journal of Behavioral Sciences, 9, 151-181. 

Oakland, T., & Shermis, M. (1989). Factor structure of the sociocultural scales. 
Journal of Psychoeducational Assessment, 7(4), 335-342. 

Oetting, E., & Beauvais, F. (1991). Orthogonal Cultural Identification Theory: 
The cultural identification of minority adolescents. The International Journal 
of the Addictions, 25(5A & 6A), 655-685. 

Padilla, A. (1980). The role of cultural awareness and ethnic loyalty in 
acculturation. In A. Padilla (Ed.), Acculturation: Theory, models and some new 
findings. Boulder, CO: Westview. 

Padilla, A., Wagatsuma, Y, & Lindholm, K. (1985). Acculturation and personality 
as predictors of stress in Japanese and Japanese Americans. Journal of Social 
Psychology, 125, 195-205. 

Padilla, A.M., & Salgado de Snyder, N. (1985). Counseling Hispanics: Strategies 
for effective intervention. In P. Pedersen (Ed.), Handbook of cross-cultural 
counseling and therapy (pp. 157-164). Westport, CT: Greenwood. 

Parham, T. (1989). Cycles of psychological nigrescence. Counseling Psychologist, 
17(2), 187-226. 

Parham, T, & Helms, J. (1985). Relation of racial identity attitudes to self- 
actualization and affective states of Black students. Journal of Counseling 
Psychology, 32(3), 432^40. 

Park, R.E. (1928). Human migration and marginal man. The American Journal of 
Sociology, 33, 881-893. 

Parsonson, K. (1987). Intermarriages: Effects on the ethnic identity of the 
offspring. Journal of Cross-Cultural Psychology, 18(3), 363-371. 


Phinney, ]., & Alipuria, L. (1990). Ethnic identity in college students from four 
ethnic groups. Journal of Adolescence, 23(2), 171-183. 

Phinney, J.S. (1992). The Multigroup Ethnic Identity Measure: A new scale for 
use with diverse groups. Journal of Adolescent Research, 7{1), 156-176. 

Pumariega, A.J. (1986). Acculturation and eating attitudes in adolescent girls: A 
comparative and correlational study. Journal of the American Academy of Child 
Psychiatry, 25{1), 276-279. 

Rauch, S., Bowler, R., & Schwarzer, R. (1987, August). Acculturation effects on 
personality characteristics in three minority groups. Paper presented at the meeting 
of the American Psychological Association, New York. 

Richman, J., Gaviria, M., Flaherty, J., Birz, S., & Wintrob, R. (1987). The process 
of acculturation: Theoretical perspectives and an empirical investigation in 
Peru. Social Science Medicine, 25{7), 839-847. 

Rueschenberg, E., & Buriel, R. (1989). Mexican- American family functioning and 
acculturation: A family systems perspective. Hispanic Journal of Behavioral 
Sciences, 11{3),232-2U. 

Ruiz, A. (1990). Ethnic identity: Crisis and resolution. Jouryial of Multicultural 
Counseling and Development, 18{\), 29-40. 

Sabogal, E, Marin, G., Otero-Sabogal, R., & Marin, B. (1987). Hispanic familism 
and acculturation: What changes and what doesn't? Hispanic Journal of 
Behavioral Sciences, 9(4), 397-412. 

Sabogal, E, Otero-Sabogal, R., Perez-Stable, E., & Marin, B. (1989). Perceived 
self-efficacy to avoid cigarette smoking and addiction: Differences between 
Hispanics and non-Hispanic Whites. Hispanic Journal of Behavioral Sciences, 
11(2), 136-147. 

Salgado de Snyder, N. (1987). Mexican immigrant women: The relationship of 
ethnic loyalty and social support to acculturative stress and depressive 
symptomatology. Spanish Speaking Mental Health Research Center Occasional 
Papers, 22, 73. 

Smelser, N. (1968). Essays in sociological explanation. Englewood Cliffs, NJ: Prentice- 

Smith, T.W (1980). Ethnic measurement and identification. Ethnicity, 7, 78-95. 

Social Science Research Council (SSRC). (1954). Acculturation: An exploratory 
formulation. American Anthropologist, 56, 973-1002, 

Sodowsky, G., Lai, E., & Plake, B. (1991). Moderating effects of sociocultural 
variables on acculturation attitudes of Hispanics and Asian Americans. /oi/nifl/ 
of Counseling Development, 70, 194-204. 

Sodowsky, G., & Plake, B. (1991). Psychometric properties of the American 
International Relations Scale. Educational and Psychological Measurement, 51, 


Spindler, L., & Spindler, G. (1967). Male and female adaptations in culture change: 

Menomini. In R. Hunt (Ed.), Personalities and cultures (pp. 56-78). New York: 
Natural History Press. 

Stephan, W., & Rosenfeld, D. (1978). Effects of desegregation on racial attitudes. 
Journal of Personality and Social Psychology, 36(8), 795-804. 

Suinn, R., Rickard-Figueroa, K., Lew, S., & Vigil, R (1987). The Suinn-Lew Self- 
Identity Acculturation Scale. Educational and Psychological Measurement, 47(2), 

Thompson, R.H. (1989). Theories of ethnicity: A critical appraisal. New York: 

Triandis, H., Kashima, Y, Shimada, E., & Villereal, M. (1986). Acculturation indices 
as a means of confirming cultural differences. International Journal of Psychology, 
21(1), 43-70. 

Trimble, J. (1988a). Multilinearity of acculturation: Person-situation interactions. 
In D. Keats, D. Munro, & L. Mann (Eds.), Heterogeneity in cross-cultural 
psychology (pp. 173-186). Berwyn, PA: Swets & Zeitlinger. 

Trimble, J. (1988b). Putting the etic to work: Applying social psychological 
principles in cross-cultural settings. In M. Bond (Ed.), The cross-cultural 
challenge to social psychology (pp. 109-121). Newbury Park, CA: Sage. 

Trimble, J. (1991). Ethnic specification, validation prospects, and the future of 
drug use research. International Journal of the Addictions, 25(2A), 149-170. 

Trimble, J.E. (in press). Toward an understanding of ethnicity, ethnic identity, 
and their relationship with drug use research. In G.J. Botvin, S. Schinke, & M. 
Orlandi (Eds.), Drug abuse prevention with multi-ethnic youth. Newberry Park, 
CA: Sage. 

Trimble, J.E., & Fleming, C. (1989). Providing counseling services for Native 
American Indians: Client, counselor and community characteristics. In P. 
Pedersen, J. Draguns, W. Lonner, & J. Trimble (Eds.), Counseling across cultures 
(3rd ed., pp. 145-168). Honolulu: University of Hawaii Press. 

Vaughan, G.M. (1986). Social change and racial identity: Issues in the use of 
picture and doll measures. Australian Journal of Psychology, 38(3), 359-370. 

Vaughn, J., Lin, N., & Kuo, W. (August, 1974). Some major determinants of stress 
among Chinese Americans: A test of two theories. Paper presented at the meeting 
of the American Sociological Association, Montreal. 

Weinreich, P. (1986). The operationalization of identity theory in racial and ethnic 
relations. In J. Rex & D. Mason (Eds.), Theories of race and ethnic relations (pp. 
299-320). Cambridge, England: Cambridge University Press. 

Westermeyer, J. (1984). The role of ethnicity in substance abuse. Advances in Alcohol 
and Substance Abuse, 4(1), 9-18. 

White, L. (1947). Culturological vs. psychological interpretations of human 
behavior. American Sociological Review, 12, 686-698. 


White, R.W. (1974). Strategies of adaptation: An attempt at systematic description. 
In G.V. Coelho, D.A. Hamburg, & J.E. Adams (Eds.), Coping and adaptation. 
New York: Basic Books. 

Wong, H.Z. (1982). Asian and Pacific Americans. In L. Snowden (Eds.), Reaching 
the Underserved: Mental health needs of neglected populations (pp. 185-204). 
Newbury Park, CA: Sage. 

Young, M., & Gardner, R. (1990). Modes of acculturation and second language 
proficiency. Canadian Journal of Behavioral Science, 22(1), 59-71. 

Yu, L. (1984). Acculturation and stress within Chinese American families, journal 
of Comparative Family Studies, 25(1), 77-94. 

Yu, L., & Harburg, E. (1980). Acculturation and stress among Chinese Americans 
in a university town. International Journal of Group Tensions, 10(1-4), 99-119. 



The History and Philosophy of 

Science In a Global 


Philip B. Vander Velde, Pli.D., and 
Hyung-Clian Kim, Ph.D. 


This chapter introduces the problems and tasks involved in un- 
dertaking a course in the philosophy of science as it pertains to a 
doctoral program in "community psychology/' It appears to us, 
the authors of this course, that this seminar has a number of deeply 
intriguing facts that are not found in either philosophy of science 
or philosophy of psychology courses. These unique theoretical 
facts, however, should provide fertile ground for developing sub- 
sequent theoretical problems and solutions in the arena of com- 
munity psychology 

The concept of "community psychology" comprises two con- 
cepts, which usually belong to separate academic disciplines and 
levels of inquiry Psychology has traditionally concerned itself with 
the inner working and motivation of individual human beings and 
has taken as its theoretical grist behavior and /or the causes of be- 
havior. Sociology, on the other hand, takes as its domain of discourse 
collections of these individuals. Both disciplines, nevertheless, use 
as their dominant mode of analysis the flesh-and-blood individual 
in the tradition of Western individualism. Usually, courses in the 


philosophy of science approach science and human inquiry in terms 
of this fundamental orientation of individualism. 

A doctoral program in community psychology takes as its area 
of inquiry two realities that can only be approached from the per- 
spective of systems theory. The program is a fundamental diver- 
gence from traditional programs in both psychology and sociology 
and assumes a welcome and daring approach to the understand- 
ing of individuals as embedded in their communities. It is these 
communities that then form the central unit of analysis, and it is 
the dynamic interplay of autonomous systems that provides the 
intellectual landscape for the course. 

Two Aspects of Science 

Science is both action and adventure. As action it involves obser- 
vation, data gathering, and analysis. As adventure, it requires 
imagination, daydreaming, and a leap of faith into the unknown 
and unexplored territory of nature and its relations with man. 
The former aspect of science is somewhat mechanical, the latter 
mystical and spiritual. Science as a mystical vision of humans 
and their exploration of nature is very much subsumed in a par- 
ticular type of perception, cognition, evaluation, and judgment, 
which humans' cultures filter for them by means of dominant 
value structures, and the cultures' belief systems and languages. 
Science, therefore, cannot be separated from a particular world 
view or cosmology of a culture where it is pursued as action or 

It is almost impossible to think of the origin of science with- 
out speculating upon its links to magic or mystical visions of early 
peoples. As Malinowski (1954) eloquently stated, there have been 
in existence in every primitive community two domains, the sa- 
cred and the profane; the former is the domain of magic and reli- 
gion, the latter that of science. Whether the object of their worship 
was the sun or the moon, or a bear or a tiger, early humans must 
have thought that those held forces beyond human control, forces 
of magic. When they observed an extraordinary event in their 
mundane world of repeated experience, they must have believed 
it was associated with forces that could not be either understood 


or explained by human reasoning. Lightning that ignited a forest 
fire or a torrential rain that swept away crops could not be ex- 
plained within the normal range of reasoning power; such forces 
had to be associated with supernatural power. Primitive people, 
therefore, believed these forces were gods, ghosts, spirits, or dead 
ancestors, among others. On the other hand, tools and methods 
for farming, fishing, hunting, or healing could not have been in- 
vented by preliterate humans without their understanding natu- 
ral processes and their regularity. This understanding was the 
beginning of science. 

Fhe Beginning of Science 

Science as a way of understanding natural process and its relation- 
ship to people probably began in the Middle East some 10,000 or 
more years ago. It started when information was gathered on plants 
to be grouped and classified and when captured animals were later 
cataloged. This type of human activity required inventing a system 
of counting. Once that was invented, it was applied to the study of 
the heavens and to constructing large monuments. The invention of 
such a system was followed by written language, which helped ad- 
vance the cause of science immensely. 

The civilization that is considered by many historians as 
mainly responsible for ushering in a ''scientific age'' for the first 
tin\e in human history was developed in Greece, where great 
thinkers formed ideas about nature, tested them, and interpreted 
their results. With the dawning of the Homeric Age, there came 
gradually many philosophers of nature one after another who 
developed their unique ideas on basic elements of nature. Thales, 
born about 624 B.C., believed that water is the basic constituent of 
all things. Anaximenes, a student of Anaximander who was born 
about 610 B.C., believed that air is the basic element from which 
all things came. Heraclitus, known for his idea of balance between 
opposites in the universe, believed that fire is the primary force 
as the agent of change. 

Given the three elements of separate Greek philosophers as- 
sociated with the basic makeup of the universe, it was probably 
inevitable that someone would integrate them into one unifying 


theory to explain the basic constituents of nature. Empedocles, 
born about 492 B.C., developed the doctrine of four elements 
(Ronan, 1983). Empedocles believed that the four elements — earth, 
air, fire, and water — are the roots of all things. He also believed 
that there are two basic forces responsible for all things in the 
universe. Of all the philosophers of nature who developed ideas 
on basic elements of the universe, the Greek Atomists are prob- 
ably the best known to us for their sophisticated understanding 
of the basic composition of matter. 

Cosmology and Science 
in Traditional China 

Early Chinese philosophers entertained a similar cosmological 
view. Particularly, the Mohists seemed to have played with an 
atomic idea, but their outlook did not survive the more rigorous 
naturalist idea of five elements. Unlike the Greek Atomists, the 
Chinese Atomists developed the five-element theory, which goes 
back to between 350 and 270 B.C., when it was systematized by 
Tsou Yen, a naturalist. According to this theory, the basic elements 
that constitute the natural world are water, fire, metal, wood, and 
earth. But these were not considered simple substances; rather, 
they were considered active principles that were believed to gov- 
ern natural processes (Scharfstein, 1974). For instance, water was 
associated with descending, while fire was related to ascending; 
wood was related to the principle of cutting and carving, while 
metal was considered the principle that operated in molding; and 
earth was related to vegetation. 

The five-element theory, when coupled with another Chinese 
explanation of the world in nature, helped to shape the fundamen- 
tal Chinese cosmological outlook that remained to a large extent 
unchanged until the introduction of Western science. The second 
cosmological view the Chinese philosophers used in their attempt 
to explain nature was the idea of two fundamental forces, the yin 
and the yang. Yin was associated with the female principle, while 
yang was related to the masculine. These principles were neither 
separate from each other nor opposed to one another since each 
complemented and was essentially a counterpart of the other. 


Fundamental to the Chinese cosmology was the notion that there 
is a multiplicity of association in nature, and every process in nature 
is explainable in terms of this fivefold constellation. It is quite fitting 
here to note that the Chinese character for ''right" or "correct'' ( T' , 
zheng) represents a five and is a method of counting by groups of 
fives. When this fivefold arrangement proved insufficient, the 
Chinese later developed a different associative thinking based 
again on the numerical concepts of 4s, 9s, and 28s. Ultimately, this 
associative thinking found its way into the / Ching, the Book of 
Changes. Scholars believe that ancient peasant omen texts gave 
birth to this strange classic, and as more interpretations of three 
ancient divinational practices were added to the texts, it devel- 
oped into an elaborate system of symbols and explanations con- 
sidered able to explain any social and natural changes (Mote, 1971). 

The I Ching is made up of 64 patterns (kua) of undivided long 
and two divided short lines in all possible combinations and per- 
mutations. Each of these patterns has a particular idea assigned 
to it, and people have traditionally believed that these kuas pro- 
vide "scientific explanation" for changes in both man and na- 
ture. Scholars are divided even today over the usefulness of the 
book. One school of thought argues that the book has served no 
purpose and that it was only full of superstitious practices for 
telling fortunes. Others have maintained that the book has de- 
terred the Chinese from developing a genuine scientific method 
of observation and experiment. And still others insist that the 
book has served a useful purpose because it has taught that change 
is not only universal but permanent. Carl Jung and Hellmut 
Wilhelm, when discussing the book, pointed out the difference 
between Western and Eastern views of cosmology and stated that 
the West looks for causality while the Chinese search for 
synchronicity (Needham, 1981). 

An inseparable part of both theories of the yin-yang and the 
five elements is the notion that the ch'i, or force, animates and 
moves nature. Ch'i was conceived of in other cultures, but the 
Chinese incorporated it into the yin-yang and five elements in 
such ways that it has become a powerful way of explaining cos- 
mic questions. The c/z'z runs through both the yin and the yang, 
as Tung Chung-shu (1797-104? B.C.) suggested, and it unites 


them into one harmonious relationship. The yin is shared by the 
yang, and each complements the other. There would be no yin 
or yang without the ch'i that exists in the universe. The ch'i con- 
stitutes the five elements and works within each element, bring- 
ing them all into unity. What is suggested here is that the Chinese 
understanding of nature is deeply rooted in its cosmological view 
that all things in nature ultimately constitute one grand harmo- 
nious and mutually complementary relationship. This concept 
was called by the philosopher Chu Hsi (A.D. 1130-1200) "the 
Great Ultimate." 

Fundamental to the understanding of the Chinese cosmologi- 
cal view is its concept of time and space. No genuine science could 
have developed independently of man's conception of time, for 
time is ultimately the measure of change. As Joseph Needham 
(1981) noted, the Chinese, being organismic naturalists, have been 
philosophically inclined to emphasize the reality and importance 
of time. Rulers of various dynasties in the long history of China 
considered time so important that they had royal astronomers 
who were officially appointed to not only study time but also to 
tell the signs of the times. Generations of astronomers left writ- 
ten records of their observations of the heavens, thus enabling 
their followers to use them as guides for their own heavenly ob- 
servations. In this sense, scientific knowledge was considered cu- 
mulative and cooperative in time. The Chinese were firm believers 
that in science, as in history, books and observations go together, 
hand in glove. 

In the ancient Hellenistic and Indian cultures, time was seen 
as being cyclical. One of the consequences of such an outlook on 
time is that it shuts out any real novelty for the future, since the 
future is already determined, the present is not unique, and all 
time is essentially past time. In contrast to this cyclical view, the 
Chinese conceived of time as being both cyclical and linear. Al- 
though at times Chinese philosophers and "scientists'' speculated 
on time and space as being subjective domains of the human 
mental map, they ultimately came to the conclusion that time is 
real and applicable to everyday human affairs. 


Time is also characterized as cyclical in both ancient and mod- 
ern China. The Chinese have applied their concept of cyclical time 
to their understanding of the life cycle. Babies are born into the 
family, thus making it possible for a new life to be linked with 
dead ancestors. A dynasty is born anew with vigor, but in time is 
replaced by another dynasty, thus partially completing the cycli- 
cal nature of a dynasty. Forces in nature work in a natural cycle in 
response to one another, and the yin and the yang move in a cycle 
to complement each other. 

Time is also linear, and both infinite and cumulative. Unlike 
the Hebraic and Christian civilizations, where time is apocalyp- 
tic, the Chinese have yet to develop an eschatological view of the 
universe, where time comes to an end and all life ceases to exist. 
Whereas the Chinese conceived of time as being cyclical, it was 
also considered part of the infinite flow of history that has no 
end. Human experience marked by time is accumulated in the 
form of personal and dynastic records, and a long list of the 
dynasties' government bureaus maintained official documents. 
Time was not only recorded, but was also kept by calendar and 
clock. Actually, timekeeping was considered a very important part 
of the proper function of Chinese government because it was 
obliged to inform its subjects of astrological and astronomical 

The concept of cyclical time of the Chinese was also inseparably 
related to their understanding of themselves in the cosmos. The uni- 
verse or cosmos in Chinese consists of two characters, yu and zhou. 
The former means "eaves"; the latter, "ridgepole." The Chinese were 
keenly aware of their ontological place in the universe that is self- 
contained, as the two characters connote linguistically. 

While comparing the Chinese concept of space with that of 
Western culture, Needham suggests that the Chinese preferred 
an intricate web of interrelatedness of all things, all influencing 
each other, while Westerners leaned heavily toward the 
conceptualization of the universe as a series of independent and 
autonomous events connected by a chain of causality. This was 
essentially the Newtonian cosmological understanding that be- 
gan to dominate Western civilization in the 17th century. 


4\ "^ifi^' 

Social Science and the Newtonian 

Since the latter part of the 19th century, Western culture has both 
been enamored of and passionately devoted to the establishment 
of science. The 20th century spawned an incredible optimism 
about the power of existing physical science, and it was expected 
by the vast majority of educated people that science with accom- 
panying technological applications would usher in a Utopia in 
the 20th century It was believed that science had the capacity to 
solve all human problems. This all was the inevitable outcome of 
the 17th century Enlightenment philosophy, which posited rea- 
son as the sole arbiter of reality and human problems. 

At the turn of the century, it was thought that there were only 
a few problems yet to be solved by physicists, and hopes were 
high that the problems of electromagnetism and the nature and 
speed of light would soon be solved. There is even evidence to 
believe that some influential academicians of that era thought that 
with the resolution of these problems there would be no more 
problems to solve. There was little reason to suspect that the re- 
search on what were thought to be the last problems of the 
Newtonian physical model of reality would give rise to a new 
physics that relegated the applicability of Newtonian physics to a 
midrange of experience that does not and cannot deal with the 
very fast or the very small. 

Although science and technology have become suspect in some 
academic circles and quantum physics and relativity theory have 
virtually replaced Newtonian physics with assumptions that are in- 
consistent with the old physics, the world view of the Enlighten- 
ment and the optimism of the latter 19th century still prevail in much 
of academia and especially in the general society. We still maintain 
the staunch belief that the paradigm of the 17th century and its con- 
comitant technology will provide solutions for contemporary prob- 
lems. This is especially true in the social sciences. Most research in 
these sciences uses the assumptions of Newtonian physics. This 
grand canyon between quantum physics, systems theory, and eco- 
logical theory on the one side and the assumption of the contempo- 


rary social sciences on the other must be addressed by a course in 
the philosophy of social science. 

We must critically question what has been taken for granted 
for the past few centuries among scholars: the assumption claim- 
ing that human rationality, as formulated in the 17th century, is 
an adequate map of our experience. We must be more demand- 
ing of research in the social sciences. We must ask what the social 
sciences can say about turmoil and siege in our metropolitan ar- 
eas and what precisely these sciences can offer to the understand- 
ing and rebuilding of communities. A further query must begin 
to examine how human communities, be they biotic or ecologi- 
cal, are embedded. We must ask ourselves how the wisdom of 
indigenous peoples can be incorporated into our scientific enter- 
prise. A topic that must be incorporated into our endeavors and 
discussed seriously in a course in philosophy of science is that of 

Newtonian Assumptions 
and Western Cosmology 

Probably the most seminal book of the past few decades in the 
philosophy of science is The Structure of Scientific Revolutions by 
Thomas Kuhn (1970), who analyzed the major pragmatic shifts 
that have taken place in the development of Western science. The 
major thrust of Kuhn's analysis is that each era of science has a 
predominant, unchallenged set of assumptions (a paradigm) about 
the nature of the data, which are used to conduct the doings of 
science during that period. These assumptions about the nature 
of reality predominate until data arise that cannot be handled by 
the dominant paradigm, and competing paradigms are advanced 
by scientists who suggest virtual heresies in their new set of as- 
sumptions. Much debate and discussion has evolved from Kuhn's 
writings, and because of this, the 17th century notion of an objec- 
tive rational science has come to be challenged. The assumptions 
of each paradigm are primitive and unchallenged within the con- 
text of the system; thus, we are left seemingly with the relativity 
of those assumptions. The Enlightenment assumption of a 


standard, stock, and undeniable objective reality has been called 
into question by both Kuhn and other contemporary theorists. 

The Enlightenment assumption of an objective, knowable re- 
ality is at the heart of most research and theoretical analysis in 
psychology and sociology However, as with most predominant 
assumptions of a cosmology, it is not considered an assumption 
but, rather, the only way to see our experience, and this has had a 
profound impact on the formulation of how we conduct science 
and understand other modes of inquiry, which are granted re- 
spect in terms of their following the respected hard sciences. 

The paradigm or cosmology that developed during the for- 
mulation of the Newtonian physics, and which subsequently pro- 
vided the theoretical model of all ensuing Western science, not 
only assumes a knowable, objective reality, but also contains a 
number of other assumptions that determine how we arrive at 
what is taken to be knowledge. In order to understand more fully 
the problems that face the contemporary social sciences, we should 
explore these assumptions. 

The primary assumption is the existence of discrete and iso- 
lated objects of perception that do not depend on the perceptual 
processes of the observer for their existence. How we go about 
perceiving should have little to do with the manner in which we 
conduct science; the objects have an independent existence of their 
own quite apart from our inquiring minds. The object of the search 
is to eliminate as much of our subjective self from the process as 
we can, thus ensuring an unimpeded view of the object(s) under 
consideration. Thus, scientific inquiry is a process of discovery 
rather than an act of creation; it is a process of mapping our ob- 
jects of investigation with conceptual modes that correspond to 
the landscape. The assumption of a knowable, objective reality 
gives rise to the notion that the universe is composed of isolated 
entities that are unrelated to each other except by forces that cause 
them to act upon each other. 


The Assumption Concerning 
Science and Scliooling 

This assumption is at the heart of our educational process, and it 
structures the experience of all students continuously through- 
out their schooling. An example of this is the common experience 
in a biology class of dissecting a frog. A typical routine is to arrive 
in class one day and be asked to dissect a frog. The frog is cut 
open, the insides are carefully noted, and the results are recorded 
in a notebook. The assumption in this process is that once the 
contents of the frog are recorded, we have then come to an under- 
standing of what the frog is, quite apart from any environment in 
which the frog might live. At the very least, the ponds frogs live 
in are little more than a backdrop for the objective frog. The pro- 
cess is sometimes referred to as reductionism and basically asserts 
that the contents of the individual entity must be essentially the 
sole basis for the consideration given to the object of inquiry. 

This particular doctrine of the objective and unrelated indi- 
vidual object has permeated the practice of all the applied sci- 
ences. The obvious examples, aside from the doctrine's 
implementation in education, are in medicine and psychiatry: a 
patient comes to the physician with a complaint, and the assump- 
tions of the Western mind come into play. The patient is asked to 
explain where it hurts, while the doctor assumes the prominence 
of the individual in determining the source of the hurt. The cause 
of the discomfort is assumed to be inside the autonomous indi- 
vidual, with little concern for the environmental context of the 
discomfort. The physician then proceeds to prescribe a treatment 
that will deal with the malfunction inside the patient. Although 
systems theory has made some inroads into the profession of psy- 
chiatry, the dominant mode of analysis is virtually the same as in 
medicine: the process is to exorcise from the individual patient what 
may be causing the aberrant behavior. 


The Assumption Concerning 

The assumption of individualism also dominates in political and 
economic theory in Western culture. All political and economic 
activity is basically reduced to the striving of the autonomous 
individual for property or other forms of personal security and 
well-being. Such striving is the sole determinant of experience, 
with the environment of other beings and nature being a some- 
what incidental backdrop. The individual is basically abstracted 
from his or her environment, in the context of most analysis, with 
preeminence being given to the human drive of acquisition. We 
are still attempting to solve our political and economic problems 
in this manner, even though the problems discussed are often 

The centrality of the assumption of the autonomous individual 
in the cosmology of our culture can best be seen in the location of 
will or the cause of action. Given any particular phenomenon 
under consideration by any of the predominant modes of inquiry, 
the cause of the activity or process is always located in the natural 
individual. When the question is posed about why the frog 
moved, the analysis proceeds in terms of the frog's internal state. 
When queried as to why a murder happens or why a beneficence 
occurred, the emotional content of the individual is usually cited 
as the source of will. Our entire legal system, with its notions of 
responsibility and punishment, is based on the unverifiable as- 
sumption of the primacy of human will. 

This central assumption has been severely criticized in the 
past few decades. Although it is not within the scope of this chap- 
ter to go into these criticisms in any great detail, it should be noted 
that major developments in theoretical systems now compete with 
the Newtonian world view. These criticisms originated from the 
inability of the cosmology to deal with the data and problems it 
generated. Criticism in the social sciences is of a like nature. It 
asks whether the major theoretical strains of psychology have 
anything to say about how we may understand the problems in 
our inner cities or whether a new set of assumptions need to form 
the basis for this analysis. 


The Assumption Concerning Nature 

Capra (1982) in his The Turning Point notes that the figures re- 
sponsible for the initial formulation of the Enlightenment para- 
digm viewed nature as adversarial and hostile. Francis Bacon's 
notion that Nature's secrets must be wrested from it and that these 
secrets could be used to dominate it certainly indicate the initial 
adversarial stance of the paradigm toward the environment. This 
can be seen in the writing of Bacon and others, that the chief cri- 
terion for judging the soundness of a theory should be whether 
the theory itself can be used to predict and control. Science was 
to be used as an instrument of power over the domain of objects 
it studied. For instance, the purpose of meteorology ultimately is 
to predict and control the weather. The basic purpose of science 
in this paradigm is to use its power as a tool against whatever it 
is studying, not just to understand, but to make interventions 
and alter the course of natural processes. Whether this is desir- 
able is open to controversy. However, Western science, as it is 
carried out, definitely makes a value statement about our rela- 
tionship to the natural environment. 

The Assumption Concerning 

The last major assumption we will deal with is anthropocentrism. 
Western science not only supplies scientists and humanity with 
power, it also elevates our species to the highest point in the chain 
of being. Our species sees itself as the ultimate in the hierarchy of 
epistemological endeavors among the species on this planet and 
degrades the understandings achieved by what are seen as lower 
forms of life. There is no room for the idea that the wolf may have 
some insights into the earth and its processes that transcend the 
insights of humans, while, at the same time, we remain rather 
callous to the loss of a significant number of species epistemolo- 
gies each year. If species extinction is challenged at all, it is gener- 
ally on the grounds that humanity may be losing a cure for a dread 


Western science also lacks much respect for the knowledge 
diversity of other cultures, races, and ethnic groups. The assump- 
tion is that we must share what we have with other groups of 
people so that they too can come to have the power over nature 
that we possess. We have exported our science, technology, and 
business with the assumption that replacing the epistemological 
awareness of indigenous peoples is progress. 

There are many other assumptions in the Western paradigm 
in regard to science, time, quantification, and perception that at 
some point in a philosophy of science course need to be analyzed. 
However, we continue to believe that the assumptions of the au- 
tonomous objects of perception, adversarial behavior, and ethno- 
centrism should form the basis of discussion for such a course. It 
is in these three areas that much thinking and critical analysis 
need to be done because these assumptions are the ones that keep 
us from understanding both biotic and human communities. 
Humans must find their place in the systems and contexts that 
give rise to and sustain their very lives, and it is only through the 
development of respect for and equity with the other members of 
these systems that life can be maintained. We must find our iden- 
tities as persons in the context of our lives as well as what goes on 
inside our skins. 


Capra, F. (1982). The turning point: Science, society and the rising culture. New York: 
Simon and Schuster. 

Kuhn, T.S. (1970). The structure of scientific revolutions. Chicago: University of 
Chicago Press. 

Malinowski, B. (1954). Magic, science, and religion and other essays. Garden City, 
NY: Doubleday. 

Mote, F.W. (1971). Intellectual foundations of China. New York: Knopf. 

Needham, J. (1981). Science in traditional China. Cambridge, MA: Harvard 
University Press. 

Ronan, C.A. (1983). The Cambridge illustrated history of the world's science. New 
York: Cambridge University Press. 

Scharfstein, B. (1974). The mind of China. New York: Basic Books. 




Ajaya, S. (1983). Psychotherapy East and West: A Unifying Paradigm. Honsdale, 
PA: The Himalayan International Institute of Yoga Science and Philosophy of 
the U.S. 

Bateson, G. (1972). Steps to an ecology of mind. New York: Chandler Publishing 

Bohm, D. (1980). Wholeness and the implicate order. London: Routledge & Kegan 

Bohm, D., & Peat, F.D. (1987). Science, order, and creativity. New York: Bantam 

Boulding, K.E. (1985). The world as a total system. Beverly Hills, CA: Sage 

Boyd, R., et al. (Eds.) (1991). T/t^ philosophy of science. Cambridge, MA: MIT Press. 

Campbell, J. (1972). Myths to live by. New York: Bantam Books. 

Campbell, J. (1986). The inner reaches of outer space: Metaphor as myth and as religion. 
New York: Alfred van der Marck. 

Chang, G.C.C. (1971). The Buddhist teaching of totality. University Park, PA: 
Pennsylvania State University Press. 

Comfort, A. (1984). Reality and empathy: Physics, mind, and science. Albany, NY: 
State University of New York Press. 

Giradot, N.J. (1983). Myth and meaning in early Taoism: The theme of chaos (hun- 
tun). Berkeley, CA: University of California Press. 

Gregory, B. (1990). Inventing reality: Physics as language. New York: Wiley and 
Wiley Science. 

Gregory, R.L. (1981). Mind in science: A history of explanations in psychology and 
physics. Cambridge, England: Cambridge University Press. 

Holbrok, B. (1981). The stone monkey: An alternative, Chinese-scientific reality. New 
York: Morrow. 

Kockelmans, J.J. (1968). Philosophy of science: The historical background. New York: 

Levison, A.B. (1974). Knowledge and society: An introduction to the philosophy of 
social sciences. New York: Bobbs-Merrill. 

Lindberg, D.C. (Ed.). (1978). Science in the middle ages. Chicago: University of 
Chicago Press. 

Lock, M.M. (1980). East Asian medicine in urban Japan. Berkeley, CA: University 
of California Press. 

Lopez, B. (1990). The rediscovery of North America. Lexington, KY: University of 
Kentucky Press. 


Nasr, S.H. (1964). An introduction to Islamic cosmological doctrines. Cambridge, 
MA: Harvard University Press. 

Nasr, S.H. (1968). Science and civilization in Islam. Cambridge, MA: Harvard 
University Press. 

Needham, J. (1962). Science and civilization in China (2 vols.). Cambridge, England: 
Cambridge University Press. 

Needham, J. (1970). Clerks and craftmen in China and the West. Cambridge, England: 
Cambridge University Press. 

Olds, L.E. (1992). Methods of interrelatedness: Towards a systems theory of psychology. 
New York: University of New York Press. 

O'Leary, D. (1949). How Greek science passed to the Arabs. London: Routledge & 
Kegan Paul. 

Roland, A. (1988). In search of self in India and Japan: Toward a cross-cultural 
psychology. Princeton, NJ: Princeton University Press. 

Rorty, R. (1979). Philosophy and the mirror of nature. Princeton, NJ: Princeton 
University Press. 

Stahl, W.H. (1962). Roman science: Origins, development, and influence to the later 
Middle Ages. Madison, WI: The University of Wisconsin Press. 

Tarnas, R. (1991). The passion of the Western mind: Understanding the ideas that have 
shaped our world view. New York: Harmony Books. 

Temple, R. (1989). The genius of China: 3,000 years of science, discovery and invention. 
New York: Touchstone Books. 

Wesson, R. (1989). Cosmos and metacosmos. La Salle, IL: Open Court. 

Weston, R.S., & Thoren, V.E. (1968). Steps in the scientific tradition: Readings in the 
history of science. New York: Wiley. 

Wilber, K. (1983). Eye to eye: A quest for the new paradigm. New York: Anchor/ 

Zukav, G. (1979). The dancing Wu Li Maters: An overview of the new physics. New 
York: William Morrow. 



Psychometrics and Culture 

Walter J. Lonner, Ph.D. 


Historical Considerations 

It has been said correctly that cross-cultural psychology (or cross- 
cultural research in general) is a method rather than a subfield of 
psychology Since the late 1960s, when the ''modern'' era of cross- 
cultural psychology began, much has been written about the ben- 
efits and uses of cross-cultural psychological research. Earlier 
(Brislin, Lonner, & Thorndike, 1973) as well as more recent re- 
views and analyses of these background factors (e.g., Triandis & 
Brislin, 1984; Segall, Dasen, Berry & Poortinga, 1990; Berry 
Poortinga, Segall, & Dasen, 1992) can be consulted for a wide- 
ranging overview of this growing emphasis in psychology This 
chapter takes the position that cross-cultural psychology is pri- 
marily a strategy in psychological research that requires a some- 
what specialized approach to the solution of methodological 
problems frequently associated with using psychometric devices 
in and across different cultural contexts. These involve such prob- 
lems as sampling and the establishment of equivalence so that proper 
bases of comparisons can be enhanced. 

These and other methodological concerns have been discussed 
elsewhere, including a fairly broad treatment of the many issues 
that one will encounter in field work (Lonner & Berry, 1986). For 
instance, in sampling, one has to decide upon (1) which cultures 
to select and why; (2) which communities in each culture to select 
and why; (3) which individuals to select and why; and (4) which 


behaviors to select and why. The reason for such concerns about 
sampling is obvious: in science, one generalizes from samples to 
the populations the samples represent. A research project, cross- 
cultural or otherwise, is only as good as the sample that has been 
drawn to represent specific target populations. 

Important as sampling is, however, this chapter focuses on 
the methodological conundrum that frequently influences the 
proper use of various psychometric instruments: the establish- 
ment of equivalence. Within this area, the primary topic will be 
psychometric or scalar equivalence. The basic problem here can be 
stated simply: if the psychometric data one has for two or more 
groups (with each group coming from different samples) are not 
equivalent, then comparisons based on the analysis of the data 
are at least flawed, if not rendered invalid or useless. The goal, of 
course, would be to develop projects and their analyses so care- 
fully that concerns about equivalence will be minimized as a source 
of unwanted variance. There are many facets to this goal, how- 
ever, and this chapter summarizes the major issues involved. 

Cross-cultural psychology has only recently come of age in 
terms of sophistication in the design, implementation, and analy- 
sis of research involving samples of individuals from two or more 
cultures. Nowadays, it is the exception rather than the rule to see 
a published study that has not taken into account the kinds of 
methodological problems that would normally render a study 
suspect, if not useless. Such was not always the case. 

Cross-cultural research has always been valued as a way to 
gather interesting and potentially very valuable data. In the early 
days (a period roughly spanning the years from 1930 to the mid- 
1960s), the typical cross-cultural project in psychology was con- 
ducted by an inquisitive person, usually someone from the United 
States who might have been on sabbatical leave. And, more often 
than not, a standard project involved taking a psychological test 
(usually of intelligence) developed in the Western world and ap- 
plying it to people from other cultures, often without giving much 
attention to its applicability and without considering how mean- 
ingful it was to people in other target cultures. Thus, "sabbatical 
opportunists" often took tests to other cultures to see how they 
did on "our" tests, scales, inventories, and the like. Frequently used 


here was what some have referred to as a ''deficit'' model. That is, 
test score differences (almost always favoring the West where the 
test was likely developed) were interpreted as some sort of deficit 
in the target population. This often resulted in what are now con- 
sidered ludicrous or at least unfair interpretations, but which were 
then cloaked in racial or ethnic superiority. Some of the most con- 
troversial and socially disruptive interpretations of test score data 
were done under the aegis of safari-type testing that, by today's 
standards, was abominably unfair and shortsighted. 

Other testing was done safari-style, often involving rather 
pristine and hard-to-reach cultures. For instance, the Australian 
psychologist Stanley Porteus took his "mazes" to many cultures 
in the South Pacific, testing numerous unacculturated groups. Such 
psychometric "island hopping" no longer occurs; but before these 
efforts faded they did, to their credit, provide some interesting 
insights and contributed extensively to cautions regarding inter- 
relationships between psychometric data and the cultures in which 
the data originated. These efforts helped to usher in the modern 
era of cross-cultural psychology 

A Major Focal Point: Etics and Ernies 

The issue of psychological or psychometric imperialism has been dis- 
cussed extensively in the context of cross-cultural methodology. 
Usually covered under the so-called etic-emic dilemma or con- 
troversy, the issue involves a centrally important question: to what 
extent are there universals (etics) in the vast world of psychology 
and its numerous constructs as opposed to the other extreme, the 
emic, which focuses on relativism? This continues to be an im- 
portant issue in the area of psychometrics and culture, for a basic 
reason that seemingly cannot be totally surmounted: any test or 
scale will obviously be developed primarily in one culture or coun- 
try, using methods of test construction as well as such essentials 
as the development and definition of constructs that are cultur- 
ally isomorphic. If a totally etic perspective were adopted — one 
might call it either universalism or absolutism — then there would 
be no problem in transporting any of these tests or scales around 
the world. 


At the other extreme, if a completely emic stance were ac- 
cepted, we would have a situation of radical cultural relativism. 
This completely indigenous position would probably render cul- 
tural contrasts meaningless and would require that psychometric 
devices be developed in each specific culture, and then only after 
each culture defined what it is that it would like to measure, if in 
fact it even wanted to be involved in some form of psychometric 
analysis. One must remember that the entire psychological test- 
ing industry is primarily an artifact of the highly psychologized 
world — that is, the highly industrialized, largely Western world 
that has provided well over 95 percent of the subjects in psycho- 
logical experiments, the norming of tests, the validation of psy- 
chological constructs, and so on. One should not be surprised, 
therefore, if people from Third World cultures show little or no 
interest in devices around which legions of psychometric special- 
ists have built their careers. 

The Istanbul and Kingston 

Largely in response to the growing use of the cross-cultural 
method, as well as growing concerns regarding the use of psy- 
chometric procedures around the world, an important and first- 
ever conference concerning the issues surrounding psychometrics 
and culture was held in 1971 in Istanbul, Turkey. Underwritten 
by the North American Treaty Organization (NATO), this confer- 
ence brought together psychologists from many different coun- 
tries. The conference was organized by Lee J. Cronbach, a U.S. 
and international authority on psychological testing, and Pieter J. 
D. Drenth, an influential Dutch psychologist. This conference re- 
sulted in the first book ever to be entirely concerned with the ad- 
aptation of "mental'' tests for use in other cultures (Cronbach & 
Drenth, 1972). An expanded reprise of this conference — essentially 
a 10th anniversary — was held at Queen's University, Kingston, 
Ontario, in August 1981. Also supported by NATO, this confer- 
ence included an expanded list of topics, which were again stimu- 
lated by intense scientific and social scrutiny involving all kinds 
of psychological assessment procedures. The editors of the vol- 


ume, S. H. Irvine and J. W. Berry, were among those who attended 
the earlier Istanbul meeting (Irvine & Berry, 1983). 

Shortly before the Kingston conference, the six- volume Hand- 
book of Cr OSS-Cultural Psychology was published (Triandis, 
1980-1981). Volume 2 (Triandis & Berry 1981) contains several 
chapters germane to the general topic of psychometrics and cul- 
ture, including an extensive and rather technical discussion of is- 
sues in methodology and theory, which primarily focuses on 
intelligence and abilities (Irvine & Carroll, 1980) and a chapter on 
projective techniques (Holtzman, 1980). Other discussions of the 
problems and issues associated with cross-cultural and cross-eth- 
nic testing and assessment, especially from an applied perspec- 
tive, such as counseling and educational placement, include 
Lonner (1990), Lonner and Ibrahim (1989), Dana (1993), and 
Paniagua (1994). 

Methodological Problems of 
Enduring Importance 

It was mentioned earlier that cross-cultural psychology is prima- 
rily a method of inquiry rather than a subfield of psychology. 
However, as a method it presents several problems, unique to 
cultural inquiry, that must be addressed. The most vexing of these 
problems involves the establishment of equivalence. It goes with- 
out saying that if valid comparisons across cultures or ethnic 
groups are to be made by using psychometric devices of any sort, 
the equivalence of measures must be made. Four types of equiva- 
lence have received considerable attention: functional, conceptual, 
linguistic, and scalar ox metric. 

]n functional equivalence, the primary concern is whether cul- 
tural artifacts, institutions, roles, behaviors, and so on, all of which 
of course vary around the world, can be placed in the same com- 
parative framework. As pointed out elsewhere, for instance, ''One 
cultural group's sign of status . . . may be a small herd of cattle, 
another group's functional counterpart may be a pickup truck and 
a chain saw; a third group's functional counterpart may be a va- 
cation to the Virgin Islands" (Lonner & Ibrahim, 1989, p. 306). 
Earlier, Przeworski and Teune (1970, p. 92) gave a colorful 


example of functional equivalence: 'Tor specific observation a 
belch is a belch and nepotism is nepotism. But within an inferen- 
tial framework a belch is an 'insult' or a 'compliment' and nepo- 
tism is 'corruption' or 'responsibility'." 

Conceptual equivalence is concerned, at the level of the indi- 
vidual, with the meaning that persons attach to specific stimuli 
such as test items. For example, while the simple word (concept) 
"hamburger" is associated, in much of the Western world, with a 
common sandwich made by placing a patty of beef in a sliced 
bun, there are vast numbers of people (Hindus and vegetarians, 
for example) who would evaluate the concept of "hamburger" 
much differently than a person from Denver or Boston or, thanks 
to McDonald's, even Moscow. Such seemingly endless complica- 
tions in meaning across cultures is the main reason why many 
believe that it is impossible to develop a truly "culture-fair" test. 
The only way that fairness might be achieved would be if all 
stimuli in a test or scale were either equally familiar or equally unfa- 
miliar to all people who might be asked to respond to the items 
on a test, scale, or inventory. No one has ever developed an iron- 
clad case for meeting such criteria. Because of this, one of the most 
central concerns in psychometrics and culture involves a variant 
of conceptual equivalence: linguistic or translation equivalence. 

There is an extensive body of literature covering this type of 
equivalence. Anyone who aims to adapt any sort of psychometric 
device for use in cultures where the language, vernacular, or gram- 
mar differ from the culture in which the device was developed 
should become thoroughly familiar with this literature. Excellent 
sources include Brislin (1976, 1980, 1986) and Sechrest, Fay, and 
Zaidi (1972). The process of hack-translation and its closely related 
topic of decentering are especially important when modifying ex- 
isting instruments for use elsewhere and for creating new items 
for multicultural use. 

An interim comment on what Sharma (1977) called the para- 
dox of equivalence poses an interesting problem. This paradox was 
summarized by Lonner and Ibrahim (1989) as follows: 

If a test is to yield comparable results across cultures to demon- 
strate equivalence, then as it becomes more equivalent the less 


likely will cultural differences be found. If, in contrast, one looks 
primarily for cultural differences and ignores the problem of 
equivalence, then the greater the likelihood of finding cultural 
differences. In the latter case, however, the differences that are 
found may reflect inadequacies in the establishment of equiva- 
lence rather than real cultural differences. 

This interesting dilemma can be stated another way: as a test 
item or some other putatively "neutral'' stimulus is transported 
to another culture, its form, function, and meaning may vary in 
potentially unpredictable ways. Note, however, that this paradox 
concerns comparisons or contrasts. If the goal is only to develop or 
adapt items or scales for use in only one new culture or linguistic 
group, one needn't be overly concerned with this paradox; per- 
haps one need not be concerned at all in such cases. 

The fourth type of equivalence has been called scalar or psy- 
chometric equivalence, and has been discussed extensively (by 
Poortinga, 1983, 1989; Malpass & Poortinga, 1986; Berry et al, 
1992). Paraphrasing Berry et al., the concern here is the extent to 
which comparisons might be misleading. Consider, for example, 
a comparison involving samples of individuals from two popula- 
tions, A and B. On the one hand, a dimension measured in culture 
A may not be the same as that {assumedly same) dimension in 
culture B. This could happen, for example, if length in A would 
be compared with weight in B. On the other hand, the scale units 
for (again assumedly) a common dimension may not be the same 
for the two groups. This could happen if length in A was mea- 
sured in centimeters and in B was measured in inches. In the first 
example, qualitatively different dimensions are being compared, and, 
in the second example, the quantitative units of measurement are not 
the same for the dimension or dimensions being compared. As stated 
in Berry et al. (p. 238), "Obviously, any meaningful comparison pre- 
sumes a qualitatively and quantitatively identical scale of measure- 
ment, even though such a scale is not observable and may not even 
be stated explicitly by the researcher who makes a comparison." 

Berry et al. (1992) go on to emphasize how necessary it is, 
again for purposes of making comparisons across groups, to make 
sharp distinctions between the scale of a concept and the scale of 
measurement. Here is their explicit example: 


Let us say we want to compare two cultural populations on some 
cognitive ability. This is a hypothetical construct for which only a 
hypothetical scale can be postulated, but this hypothetical scale 
is the actual comparison scale we are interested in. The instru- 
ment that provides the data forms a measurement scale. Thus, 
in sense we are dealing with two scales. One is the scale of the 
theoretical concept. The other is the measurement scale; in our 
example this is the scale of the ability test score variable. When- 
ever an instrument has validity for a concept this can be thought 
of as a mathematical function to express the relationship be- 
tween the two scales. Data are incomparable or unequivalent 
when the function expressing this relationship is not the same in 
the cultures to be compared. 

Marsella (1987) gave an interesting applied example of this 
probleni. While studying ''life satisfaction" in the Philippines, he 
asked subjects to rate how satisfied they are on the basis of a 
"simple'' scale based on five steps. He told one subject the follow- 
ing: "Juan, the man at the top of the stairs is very satisfied with 
his life, while the man at the bottom of the stairs is very dissatis- 
fied with his life. On what step would you place yourself, Juan?" 
Juan thought about his answer, and then finally said that he would 
place himself on the bottom stair. This seemed to be accurate and 
to validate Juan's situation, which involved living in a slum. But 
when Marsella asked Juan to say why he placed himself on the 
bottom step, Juan replied, "Because I do not wish to fall down the 
stairs and hurt myself. Doctor" (p. 388). Note in this example that 
being a Filipino was not necessarily the basic reason for this deci- 
sion. However, such scaling techniques may be much more diffi- 
cult to use with people who are not accustomed to such common 
scaling techniques. Resolving matters associated with psychomet- 
ric equivalence is not easy, as the above example attests. Some- 
what paradoxically, absolute scale identity or equivalence may 
never be established, even in intracultural psychological research. 
For instance, how can one be absolutely certain that acid tests of 
equivalence, especially at the quantitative or scale level, have been 
established when one is trying to assess connections between per- 
formance on an ability test and scores on a scale of test anxiety? 
What seems necessary here is to develop a nomological network 


involving hypothetical constructs and the validation of those con- 
structs (Cronbach & Meehl, 1955). 

Confidence in the measures that have been used across cul- 
tures can be enhanced by developing a nomological network. Simi- 
larly, confidence in the meaning of a single test can be improved 
by using additional, supplementary information to bolster one's 
case. For instance. Berry (1979) used the Asch procedure (Figure 
1) as an estimate of conformity behavior when he contrasted a 
group of hunters (Eskimos) with a group of agriculturalists 
(Temne) in Africa. In the Asch procedure, several confederates of 
the experimenter choose one of the lines that is clearly and objec- 
tively wrong (for instance, the confederates might say that alter- 
native 4 is closest in length to target line A); on cue, the "wrong" 
choice is witnessed one at a time by the participants in the experi- 
mental group. The dependent variable in the study was the num- 
ber of times agriculturalists (who were hypothesized to be more 
conforming) ignored confederates' choices and answered accord- 
ing to their own estimates. 

As predicted, the hunters were more independent in their 
choices than were the agriculturalists. That is, they went along 
with their confederates' choices far less frequently than did the 
agriculturalists. In making the claim that Eskimos are therefore 

Figure 1. Typical Stimuli in the Asch Procedure. 


more independent than the Temne, Berry is assuming metric 
equivalence. He is assuming that if one person in the study re- 
ports that line 4 is the same length as the standard line, then all 
persons in the study would view the length of the lines and the 
spacing between them the same way If Berry had stopped right 
there, the results of the study would have been ambiguous; one 
could in fact have argued that the results could be explained by 
inequivalence of the stimuli between groups. In other words, 
measurement error as a result of nonequivalent stimuli might ac- 
count for the difference, thus nullifying what could have been 
announced as a substantial finding. However, Berry brought in 
additional evidence to bolster his case. He presented other mea- 
sures of conformity-nonconformity He linked certain child- 
rearing practices in both places with conformity behavior. He ana- 
lyzed the food accumulation practices of the two groups, present- 
ing detailed anthropological documentation that suggested that 
the groups really do differ on conformity behavior. That his entire 
package of results was consistent and formed a clear pattern gave 
Berry confidence that the Asch procedure indeed measured con- 
formity behavior and was therefore a metrically equivalent pro- 
cedure. In experiments of this kind across cultures — indeed, in 
any kind of data-gathering procedure involving individuals from 
different cultures — multiple measures are very important. A single 
measure might be so ambiguous as to defy clear interpretations 
(or to invite faulty interpretations). 

Types of Scales Used in Psychological 

Against the background of the problems associated with scalar or 
metric equivalence, it is worthwhile to review briefly the four kinds 
of scales that are used in psychological research. In nominal scales, 
one needs only to name or label phenomena for the purpose of 
sorting people into mutually exclusive categories. Sex, ethnicity, 
religion, and political affiliation are examples. Numbers could be 
assigned to each category (male = 1, female = 2, for instance), but 
such numbering is purely arbitrary; numbers assigned in this way 
have none of the properties usually associated with numbers, such 


as ranking or magnitude. With nominal scales one cannot per- 
form arithmetic operations such as adding, subtracting, multiply- 
ing, or dividing. 

In ordinal scales or measures, some inherent order is defined 
and quantified. Level of income is a good example of ordinal scal- 
ing because, in addition to each of the levels being mutually ex- 
clusive, there is a fixed order. Thus we can speak of a given 
category as ranking higher or lower than some other category. 
Arithmetic operations are still not possible with this higher level 
of scaling; the only characteristic ordinal scales have that nomi- 
nal do not is the fixed order of the categories. 

The next highest level in measurement is interval scaling. In- 
terval measures have mutually exclusive categories and an inher- 
ent order, like ordinal measures, but they also have equal spacing 
between the categories. Temperature scales are true interval scales. 
There is exactly the same difference of 10 degrees between tem- 
peratures of 10 and 20 and between temperatures of 30 and 40. 
Another characteristic of interval scales is that the point on the 
scale labeled zero is arbitrarily selected. In temperature scales, for 
example, neither degree Centrigrade nor degree Fahrenheit is 
absolute 0, the complete absence of heat. Because of this arbitrary 
zero point, we cannot perform ratio functions. We cannot say, for 
example, that a temperature of 25 degrees is exactly twice as cold 
as a temperature of 50 degrees. Despite this shortcoming, with 
interval scales we can perform the basic arithmetic functions. 

It is at this level of psychological scaling that considerable con- 
troversy exists. Only a few measures in psychological measure- 
ment seem to qualify as interval scales. One of them involves 
intelligence testing as traditionally defined and practiced. Thus, 
although it is accurate (the scales were constructed to be so) to 
say that an IQ of 100 is 10 points higher than an IQ of 90, it is not 
correct to say that someone with an IQ of 120 is half again as intel- 
ligent as someone with an IQ of 80. It can be argued that early 
psychologists, such as Wundt and Fechner and their followers, 
made a critical mistake when they started the field of 
psychophysical scaling as if the vicissitudes of human thought 
and behavior could be placed on scales as accurate as those to 
measure the relatively unchanging qualities of the nonhuman 


world. It has often been joked that psychology has suffered from 
"physics envy/' 

Anyone interested in the highest level of measurement aspires 
to capture the qualities of ratio scales or measures. Such scales 
have all the characteristics of interval scales, but the zero point is 
absolute rather than arbitrary Most important, they permit state- 
ments about proportions and ratios in addition to permitting all 
the standard mathematical operations. If a scale used in cross- 
cultural research passed the test of scalar or metric equivalence, 
there would be little question that the scale met or approached 
the criteria required of ratio scales. The vexing problem, however, 
remains: metric equivalence has to be established rather than as- 
sumed. It is inherent in cross-cultural research that cultures vary, 
and vary to an unknown extent, on many dimensions of human 
behavior. There appears to be no lingua franca in the psychomet- 
ric world that would instill total confidence for cross-cultural use. 
And this is not necessarily because cross-cultural research offers 
its own set of specialized problems; it is simply difficult, in deal- 
ing with the vagaries of psychological constructs and the fluid 
nature of psychological terms, to establish firm bases of compari- 
son using quantitative methods. Analysis of variables across cul- 
tures just compounds an already difficult problem. 

Standardized Assessment 

The domain of psychological testing and assessment has cus- 
tomarily been divided into three broad areas: behavioral, objective, 
and projective. Each has its own tradition, and each has both ad- 
vantages and disadvantages in the cross-cultural realm. A brief 
consideration of the main characteristics of each of these areas is 
presented in this section. Table 1 compares these three major as- 
sessment approaches. 

Important distinctions to keep in mind are that behavioral tech- 
niques look at the obtained data as a sample of behavior, objective 
techniques see the obtained psychometric data as a correlate of 
something else (usually nontest phenomena), and projective tech- 
niques assume that the data are indications of a sign of some un- 
derlying personality structure or inner dynamic. At the level of 












O jz 


oi c 

O ru 

1/1 ._ 

- ^ 
^ CT 

in o '^ 


O ^ n3 
75 O M= 

oj .2 




2^ -^ LT) 

O Si 


O t>o 

S O 

c ■> 

O) 03 

=3 JZ 

cr CD 

03 Q- 

O u 
Qj QJ-O 


O =5 



. o 
c -Q 

9- Q- 


-a E 

CD _Q 

o 2 





tj on 

.;= o 

O ^ 

u E 



CD — 
• 03 

Q. OI 

O > 05 

■- JZ -Q 

00 T^ i_ 

3 O <U 

O _Q > 

=5 > S 

O^i^ O 


< to 03 

o <i> 

03 oo 

=3 <LI 

CD 03 


3 03 C 


CD 00 

> CD 

S o 

O U 

(U ^ 

U ;/i- 

? ^ 

-c: CD - 

U .t^ U 



8 ^ 

. c 

00 13 

t o 

O u 

75 ^ 



00 O 

o a^ 


03 X3 
"D cu 
— U 

03 3 

O. O 





C ^ 

^ (l* 

03 +-• 

^ 03 

-Q CD 

O i: 


1/1 ^ 




















oo CD 

03 ^ 

S^ _o 


03 . 

Q- 5 
CD iZ 
U Ll_ 

Q- CD 

^ O ^ 

>^jc O 
03 -^ ct; 

^ 8^ 

CU ^ 
03 oo" 

^ E 

00 "'"' 

O U 

^. o 

00 -CI 
I— Q- 




c o 


c -o 

03 CD 



C ^ 


03 CT) 

U 03 

M= 13 

00 cn 

00 C 

03 no 

U : 

03 s^ 

03 r= 
-C Z3 

X _- 

u +^ 
d 03 

O u 

00 )=;. 
CL 2 

CD _> 
■»- JZ 


CD -43 

, CV fX3 

< 03 5 
i= 03 >, 

U Jr 03 

>^ E — 

OJ o o 
■^ u > 

03 CD -^ 
> u ^~ 

ro 3 -O 
CD c C 


03 _^ 

03 OQ ^ 

o ^^ ^ 

'+->. CI " 


23 ^ v^ 
M- 1/1 LL 

— U 
03 .p 
Q- CD 

^ o 

Q_ :f3 Z3 






















00 03 Q_ 

-'^ 00 _~ 
C l/^ 

ro "D Q_ 

c^ 03 UJ 

UO Z3 

O Ln 
^ >, 

> c/^ 

03 ^ 

-C 03 



b ^ i^ 

^ S S 

Q. t E 


subjective interpretation, behavioral techniques are low, objective 
techniques are medium, and projective techniques are high. While 
this is generally the case in monocultural research, it is usually 
much more so in cross-cultural research. In cross-cultural research, 
especially that dealing with children, there is considerable litera- 
ture detailing the procedures to be used in the systematic obser- 
vation of children in naturalistic settings (Longabaugh, 1980; 
Munroe & Munroe, 1993). Indeed, the six-culture study (Whiting 
& Whiting, 1975) relied heavily on standardized systematic ob- 
servation of children and their interrelations with others for the 
purpose of attempting to determine which factors, if any, account 
for patterns in human behavior. 

Behavioral techniques involve more than simply observing 
and recording behavior on checklists. They also involve the con- 
struction of tasks and procedures that are designed to allow the 
researcher to observe how the individual approaches possible so- 
lutions. For instance, the clinical method developed by Jean Piaget 
was designed to elicit samples of cognitive maneuvers regarding 
thought processes and approaches to problem solving, while keep- 
ing levels of inference quite low. Piagetian assessment, however, 
belongs in a later section concerning nonstandardized assessment. 
Consequently, a more complete discussion of Piagetian approaches 
is included there. 

A large percentage of cross-cultural research involves the use 
of any number of thousands of available tests, scales, question- 
naires, and inventories that are assumed to be objective, and to 
possess the necessary conditions pertaining to their use — in par- 
ticular reliability and validity. Following the dichotomy provided 
by Cronbach (1990), objective tests can be listed as either tests of 
maximum response or tests of typical response. 

Tests of Maximum Response 

Measures in this broad category include those that try to assess 
how well a person can do, how much "mental capacity'' he or she 
has, or generally how well he or she can perform when subjected 
to the rigors of such tests. The best example of this type of device 
is the well-known and controversial intelligence test. The various 


Wechsler scales, the Stanford-Binet, Raven's Progressive Matri- 
ces, and many others have been involved in seemingly countless 
attempts to understand relationships between culture and com- 
petence. In cross-cultural psychological research, and in psycho- 
logical anthropology as well, there has indeed been a relentless 
pursuit to understand the role culture plays in the cultivation of 
human abilities. There have been three approaches to understand- 
ing this relationship: the general abilities approach, the specific abili- 
ties approach, and the cognitive styles approach. 

The General Abilities Approacli 

Psychologists who have used this approach usually made two 
assumptions. First, any respectable intelligence test — especially 
those that are supposedly saturated with "g," or the general abili- 
ties factor in the spirit of Charles Spearman (1904) — need only be 
translated for use in cultures for which it was not originally in- 
tended. Second, it follows from the first assumption that cultures 
can form a continuum from high to low in terms of general ability 
in the same sense that individuals can be arranged from high to 
low with the same test. The widespread earlier use of Raven's 
Progressive Matrices was driven by this approach. Even though 
the idea of ''g" still has popular appeal, especially to those who 
take a factor analytic approach in measuring human abilities, this 
general abilities approach is no longer very popular in compara- 
tive cultural analysis. A main reason it has lost its earlier follow- 
ing is because it is essentially ethnocentric. 

The Specific Abilities Approach 

Partly in response to charges of ethnocentrism, and partly because 
there is still lingering doubt about what intelligence or compe- 
tence ''is'' let alone how it can be measured accurately, a number 
of psychologists have looked into specific abilities possessed by 
people in different cultures. Consistent with the anthropological 
idea of the "psychic unity of mankind," there have been searches 
for abilities that are emic — that is, meaningful only in specific 
cultural contexts. The question asked by people who have used 
the general-abilities approach is, "How smart are they?" 


(as measured against a common yardstick provided by a ready- 
made test). Contextualists (Neimeyer, 1993) turn this question 
around and ask, ''How are they smart?'' 

Those who take the contextualist approach will likely not use 
any ready-made psychometric device, such as a Wechsler scale, 
even if it is competently translated. Rather, they will typically care- 
fully examine the unique essence of a culture, much like a cul- 
tural anthropologist will function in the process of "deep cultural 
immersion." In fact, this approach has been called "experimental 
anthropology" and "unorthodox psychology" — terms that its 
adherents might readily accept. The person who has been most 
closely associated with this approach is Michael Cole (Cole, Gay, 
Click, & Sharp, 1971; Cole & Scribner, 1974). This approach has 
great appeal. Most psychologists who study relationships between 
culture and abilities subscribe to the basic logic of this approach 
primarily because of the belief that any attempt to measure intel- 
ligence or abilities must take into account indigenous cognition. 
Serpell (1993) supports this view by arguing that intelligence is 
culturally constructed and consists of three basic elements: 

• Culture as a nurturant environment. Every culture "feeds the 
growing mind and channels children's exploratory activity 
in certain directions by specifying the material environment 
to which the child has access." In this sense, culture is 
analogous to a womb. 

• Culture as a system of meanings. Serpell reminds us that 
"fairness and the capacity to appreciate another person's 
point of view often carry great weight" when we are seek- 
ing advisors in moments of crisis. These are qualities of 
mental depth rather than speed, and they play an important 
role in developing the goals of childrearing in many indig- 
enous African cultures. The end-points of goals in these 
societies is construed as someone who can "preside effec- 
tively over the settlement of a dispute, whose judgment can 
be trusted in questions of character, and who can and will 
take on social responsibility." Needless to say, this sort of 
formulation lies beyond traditional and "orthodox" psycho- 
metrics, and would be extremely difficult to put on a com- 
mon yardstick for multicultural comparisons. 


• Culture as a forum. In many cultures, there can be different 
perspectives on the value of formal schooling compared 
with the kind of informal schooling that a child might learn 
at home. When such confusion exists, the culture can be- 
come a forum for the "articulation and debate of different 
perspectives on the significance of schooling as a resource 
for enhancement or frustration in the lives of local people." 
What this viewpoint means to the psychometrist is that he 
or she will likely have to be creative in attempts to measure 
abilities in cultural context. For example, Irwin, Schaefer, 
and Feiden (1974) compared the performance of two groups 
of adult men with different cultural experience on two 
versions of a cognitive task often regarded in Western 
psychology as an index of intelligence. Mano rice farmers in 
Liberia were able to sort bowls of rice according to three 
alternative dimensions significantly more often than under- 
graduate students in the United States, whereas the same 
Americans were able to perform significantly more sorts 
than the Liberians when the materials were cards displaying 
colored geometrical shapes. This study showed that cogni- 
tive flexibility in classification was found to depend on the 
cultural familiarity of the medium in which it was assessed. 
I like to refer to this study as ''turnabout is fair play,'' and as 
a practical, applied counterpart to the numerous "culture- 
relative'' tests (e.g., the Chitlins Test) that were developed to 
drive home a social and political point: any group could be 
made to look stupid or less than "intelligent" if that group is 
tested using unfamiliar stimuli or other unfamiliar methods 
that are common elsewhere. 

The Cognitive Sfyies Approacli 

For about a decade, considerable research in cross-cultural psy- 
chology was fueled by the late Herman Witkin's theory of psy- 
chological differentiation (Witkin & Goodenough, 1981). The 
essential idea in this framework is that different influences — ^bio- 
logical, social, and ecological — foster the development of differ- 
ent "cognitive styles." Anchoring the end-points of a continuum 


of cognitive styles are those who have a more ''field-dependent'' 
cognitive style and at the other end are those who are more "field 
independent." Research on cognitive styles was conducted in a 
variety of cultures, and always involved tests such as the Embed- 
ded Figures Test and the Portable Rod-and-Frame Test. A com- 
prehensive review of the cognitive styles approach appeared 
nearly 20 years ago (Witkin & Berry, 1975), just a few years before 
Witkin died. Since then, research on relationships between cogni- 
tive style and culture has declined sharply. 

One of the underlying ideas in this area had considerable 
popular appeal: that there is a patterning of cognitive styles. That 
is, a constellation of antecedent factors lead to the development 
of adaptations to these factors. The more similar the antecedent 
factors are for different cultures, the more similar will be their 
"styles." Using this rationale, there will be more similarity be- 
tween, for instance, nomadic Eskimos in northern Canada and 
Lapplanders in northern Finland than there will be between Eski- 
mos or Lapplanders and, say, sedentary groups in sub-Saharan 
Africa. John Berry and his colleagues, who substantially colonial- 
ized Witkin's theory by systematically sampling cultures to test 
hypotheses regarding adaptation, have made this approach analo- 
gous to collecting butterflies. That is, cultures or perhaps certain 
groups within cultures are selected because it has been determined 
that they meet certain criteria necessary to test hypotheses gener- 
ated by the theory. For instance, a group of sedentary rice farmers 
with a certain type of family structure (a type of group not previ- 
ously selected) might satisfy specific requirements in order to test 
the theory. The findings from the study might then help develop 
a more complete picture regarding how cultural and ecological 
conditions affect psychological functioning, just as a particular 
butterfly will help the butterfly collector complete a certain pat- 
tern. This kind of thinking even gave rise to the development of 
the idea of "sensotypes" (Wober, 1966). This idea, which has not 
been pursued in any systematic way, has its roots in the largely 
contextualistic notion that different cultures, by dint of the differ- 
ent factors that affect individuals within them, can be character- 
ized as having different sensotypes — that is, certain of their senses 
will be activated more than others. 


Experimentation Witli a Possible 
Compromise: Tine SOIVIPA 

Jane Mercer, a sociologist, was instrumental in developing the 
System of Multicultural Pluralistic Assessment (SOMPA). The 
SOMPA offered somewhat of a compromise between the tradi- 
tional psychometric approach (the "general-abilities'' approach) 
and the convincing arguments underlying the contextualist ap- 
proach. The SOMPA used the Wechsler Intelligence Scale for Chil- 
dren in the traditional way, but it also incorporated sociocultural 
norms on educational and family factors, and involved medical 
workups as well. 

Mercer has also been an outspoken critic of "business-as- 
usuar' psychometrics involving the assessment of intelligence. 
Basing the essence of her argument on the fallout of various court 
cases challenging educational decisions and possible 
misclassification (e.g., the famous Larry P. v. Riles case in Califor- 
nia), Mercer is an advocate of what she has called the pluralistic 
IQ paradigm. There are several crucial elements in this paradigm. 
Chief among them are the following: 

• IQ tests do not measure intelligence. There is no known 
technology for measuring intelligence, "mental abilities,'' 
"aptitudes," "mental potential," or related constructs. There 
is no way, operationally, to distinguish intelligence A (the 
genotype) from intelligence B (the phenotype), from intelli- 
gence C (the specific test score). (Breaking down intelligence 
into A, B, and C is attributable to Vernon, 1969.) 
The IQ test score is an index of what a person has learned. 
Mercer even takes on the vaunted idea of "g." She maintains 
that the g factor yielded by factor analysis of a wide variety 
of tests cannot be interpreted as evidence for the existence 
of "intelligence" as an entity separate from learning. The g 
factor simply means, she says, that the materials in the tests 
are drawn from the same cultural pool and that an indi- 
vidual who has learned a lot about one aspect of a culture is 
likely to have learned a lot about other aspects of that 


• Intelligence and achievement may be separable conceptu- 
ally, but they cannot be distinguished operationally She 
contends that the Stanford-Binet and the Wechsler scales are 
simply individually administered achievement tests. 

• Individually administered achievement tests, such as the 
Wechsler scales, can reliably measure what a person has 
learned about the language and cultural materials included 
in the test. Scores on such tests, she maintains, can validly 
be used to make first order inferences about the content of an 
individuars knowledge and level of proficiency in the 
language and skills covered in the test. Scores earned by 
different individuals on such tests can be compared, and 
first order inferences can be drawn that a person with an 
appreciably higher score on the test has learned more than a 
person with a lower score. 

• Inferences about intelligence are second order inferences. The 
central issue in determining the validity of using achieve- 
ment tests for inferring intelligence lies in the extent to 
which second order inferences can properly be made in a 
particular case. The measurement theory underlying the 
traditional IQ paradigm fails to distinguish between first 
and second order inferences. The pluralistic model sees the 
distinction as crucial. 

Anumber of other factors have been advanced by Mercer (1988). 
In the same forum where Mercer summarized her views concerning 
the "death of the IQ paradigm,'' Thomdike (1988) offered a com- 
pletely different view. He argued that Mercer 's ideas are guided prin- 
cipally by social policy and courts of law. Taking a traditional and 
logical positivist scientific position, Thomdike argued that there are 
two basic premises by which science operates. The first is that there 
is a universe outside of the human mind that exists independently of 
our awareness of it. The second is that there is a set of principles, at 
least in theory discoverable, by which that universe operates, and 
that these principles do not depend on our belief systems. 

Agreeing with Mercer that every society has its own defini- 
tion of knowledge and of what is acceptable, Thorndike tells of 
the Lugbara of Uganda. That tribe once had a traditional belief 


that the world outside its immediate territory was inverted. People 
and objects that were not Lugbara were thought to be upside 
down. As contact with Europeans and others increased, however, 
the Lugbara had to introduce additional complexities into their 
belief system in order to account for the empirical fact that visi- 
tors were not inverted. Later, when they ventured outside their 
immediate territory, the Lugbara had to discard the entire sys- 
tem. Thorndike uses this example to point out that it is an article 
of scientific faith that there is a lawful (in other than the judicial 
sense) external world that operates independently of our beliefs 
about it. 

Arguments persist about what intelligence ''is'' and, once de- 
fined or hypothesized, how it should be measured. Recent devel- 
opments, including Sternberg's Triarchic Theory (Sternberg, 1985) 
and Gardner's (1983) liberalization of intelligence to accommo- 
date such notions as interpersonal and intrapersonal intelligence, 
are interesting to study. However, they need to be tested in the 
crucible of culture-comparative analysis before they can be touted 
as new revelations about the nature of human intelligence. 

Tests of Typical Response 

In using tests (which is probably a misnomer in such cases), scales, 
questionnaires, and inventories in this category of measurement, 
the desire is to assess attitudes, personality traits, values, pres- 
ence of psychopathology, and many other dimensions of the per- 
son as they normally occur in everyday settings. When assessing 
in this domain, the researcher or practitioner desires to get a rep- 
resentative sample of behavior (or beliefs, attitudes, personal dis- 
positions, and so on) that may correlate with nontest criteria. Such 
measures are too numerous to mention here — even when their 
total number is diminished by counting only those that have been 
used cross-culturally However, some of these psychometric de- 
vices have been used more than others; they and their problems 
can serve as prototypes for nearly all the others. 

Before considering any selected individual measure, the fol- 
lowing questions should be kept in mind: 


• What is the nature of the construct or structure that the 
device purports to measure? Is it reasonable to believe 
that it measures some universal dimension or dimensions 
of human behavior rather than something that is culture- 
relative only? For example, many believe that anxiety or 
depression qualify as characteristics that all humans 
possess to some degree. Thus, a device designed to 
measure these phenomena across cultures would have to 
be sensitive to how accurately it measures them rather than 
determining whether it exists in one culture but not in 
others. This is basically a problem of functional equivalence. 

• How easy would it be to translate (or how accurate are 
existing translations)? With few exceptions, measures of 
typical performance involve single words or short sen- 
tences, most often couched in self-report or self-reflexive 
terms. This area, of course, focuses on conceptual or 
linguistic equivalence. 

• How might the scores or data derived from the measure 
fit in with other data or information available on a spe- 
cific culture for a specific concept? Similarly, how might 
the scores or data fit with ostensibly equivalent psycho- 
metric reference points available from other cultures for 
the purpose of making comparisons? This involves the 
problem of metric or scalar equivalence. It also involves 
the process of construct validation and the development 
of a nomological net. 

Prototypical Examples of Measures 
of Typical Response 

The Minnesota Multiphasic Personality Inventory (MMPI) pro- 
vides an excellent example of a device that invites the solution to 
problems of equivalence before it should be used cross-culturally 
The MMPI is easily the most widely used paper-and-pencil mea- 
sure of personality /pathology ever developed. Its rich and color- 
ful history, its association with the medical model of disease 
etiology, and its reliance on criteria-referenced norms have made 
it an inviting candidate for widespread use in mainly clinical set- 


tings throughout the Western world. Many excellent books de- 
tailing the development and use of the MMPI are readily avail- 
able, including those that have outlined the large revision resulting 
in MMPI-2 (Butcher, 1990; Graham, 1990; Greene, 1991). 

As discussed more extensively elsewhere (Dana, 1993; Lonner 
& Ibrahim, 1989), in some ways the MMPI is inappropriate for 
use in different cultures or ethnic groups. For instance, it started 
with a Western system of psychiatric classification (thus possibly 
a huge imposed etic); its original criterion groups came from small 
and mostly rural samples of patients in and visitors to the Uni- 
versity of Minnesota hospital about 45 years ago; it requires about 
a sixth-grade reading level, thus eliminating many people, who 
might be more typical of a culture than their literate compatriots, 
as candidates to take it; it is subject to several problems tradition- 
ally associated with self-report personality measures (although 
built-in "validity'' scales are supposed to minimize such prob- 
lems). On the more positive side, there are several reasons why 
the MMPI can be considered a respectable candidate for cross- 
cultural use. One reason is that there appears to be substantial 
universal ''core'' characteristics in at least certain kinds of psy- 
chopathology (depression and schizophrenia, for example). An- 
other attractive reason is that there is vast literature associated 
with both its construction and use; this gives the researcher and 
practitioner alike much useful background material, including 
quite a lot of useful cross-cultural material (about 125 translations 
of the MMPI are reportedly in use today). A good and defensible 
reason for its cross-cultural employment is its potential for usewithin 
a culture through the development of appropriate norms by using 
the empirical (criterion-referenced) method of scale construction. 

Details about the international use of the MMPI are given in 
Butcher and Pancheri (1976), and Butcher (1984). Dahlstrom, Lachar, 
and Dahlstrom (1986) remain a solid source of information about its 
use with ethnic groups within the United States. These sources should 
be consulted in conjunction with more recent books pertaining to 
the use of the revised MMPI (e.g.. Butcher & Williams, 1992). 

Another candidate for research on personological variables 
and their differences across cultures and ethnic groups is the Cali- 
fornia Psychological Inventory (CPI), sometimes called the 


"healthy person's MMPF' (Gough, 1987). The development of the 
CPI was indeed patterned in many ways after the MMPL An at- 
tractive feature of the CPI is that when the items were developed 
they were based on "folk concepts" — patterns of interpersonal 
interaction that occur everywhere. If this claim is true, the CPI 
should be easy to translate. Like the MMPI, this inventory has 
enjoyed considerable cross-cultural research. 

Psychometrists working with many different cultures or eth- 
nic groups in attempts to measure personality variables will have 
a much easier job if they start with a construct that has universal 
meaning and perhaps universal correlates of behavior as well. Of 
course, that is a major point that separates the radical relativists 
from the universalists or reductionists. Somewhat paradoxically, 
one may never know for certain that a particular dimension of 
personality or a specific category of psychopathology is truly uni- 
versal; there are too many cultures in the world to develop a lead- 
pipe cinch of absolute commonality. On the other hand, relativists 
will always be able to stake a claim in support of their position, 
for even extremely small mean differences across cultures can be 
used in claims of uniqueness. In this sense, all human behavior is 
essentially unique, as the existentialists and phenomenologists 
have been telling us for years. Nevertheless, there are a number 
of personological dimensions that seem to qualify, both empiri- 
cally and by dint of common sense, for the status of universality. 
Among them are anxiety, control and depression. 

Anxiety has been studied extensively across cultures, prima- 
rily through Spielberger's State-Trait Anxiety Inventory (STAI). 
The STAI purports to measure situation- or state-induced anxiety 
(a temporary state that anyone can experience) and anxiety as a 
general and enduring characteristic of the individual. Both state 
and trait anxiety are measured by using 20 brief statements for 
each, to which the individual responds. 

The STAI has been used extensively across cultures 
(Spielberger, 1984). Research with the STAI is apparently interest- 
ing and deceptively easy: translate the scale into a specific lan- 
guage, administer it to a sample of people who speak that 
language, and then determine how the scores relate to such vari- 
ables as school or work performance, or determine its factor struc- 


ture. What is usually missing in these "static group comparisons" 
are discussions ofemic or indigenous concepts of anxiety (or other 
mental states that are semantically connected to anxiety). 

As explained elsewhere, a serious psychometric problem is 
bound to arise whenever a scale such as the STAI is translated, 
administered to people with diverse cultural /linguistic back- 
grounds, and its absolute scale values (scores) used for direct com- 
parison with scale values from the culture of origin or from 
another culture. This problem is a good example of scalar or met- 
ric equivalence, and was summarized by Lonner and Ibrahim 
(1989) as follows: 

The problem is that a way to calibrate anxiety scale values with 
absolute levels of anxiety has not yet been found. If compari- 
sons of scale scores are restricted to groups within a new culture 
or ethnic community, there would be no problem. But if com- 
parisons are to be made across cultures, the absence of an ab- 
solute (common) reference scale presents difficulties. For tests 
or scales to be directly interpretable on mean scale values across 
culture/linguistic groups, several things must be achieved: (1) 
each item's validity as an indicator of the underlying trait should 
remain unchanged; (2) the scale values and probability of en- 
dorsement at the item level should be the same; and (3) all the 
items taken together should have the same average validity, the 
same average absolute scale value, and the same operating char- 
acteristics in all cultures to which the inventory is extended. It is 
most unlikely that all of these conditions can be met in an abso- 
lute sense. 

The STAI is not singled out here as the only measurement 
device that should meet these stringent requirements. Exactly the 
same situation exists with all other measures of personality, pa- 
thology, attitudes, and so on. Psychology rarely deals with a world 
of stable and immutable phenomena that can confidently be ana- 
lyzed by employing ratio scales and the assumptions underlying 
their proper use. Human behavior is extremely complex and 
multidimensional, all but defying measurement at levels of exac- 
titude appreciated by physicists, chemists, and others who have 
the pleasure (or the boredom) of working with worlds that are 
much less dynamic and variable than the human mind. 


Yet another construct with seemingly international appeal 
because of its centrality in human behavior is control: the extent to 
which one can predict one's lot in life amid what is for many a 
bubbling, buzzing, and capricious world. For nearly 30 years the 
idea of ''locus of control" has been a fixture in social and person- 
ality psychology The chief focal point here, in terms of measure- 
ment, is the I-E Scale. This scale purports to measure how 
"internar' or "external" a person is; the more internal a person is, 
the more he or she has control of events that impinge on one's 
life. Conversely, the more external a person is the more he or she 
believes that luck, fate, chance, acts of God, and so on are respon- 
sible for one's lot in life. This concept and how it should or should 
not be measured has stimulated much research and even more 
controversy A valuable review of the locus of control concept from 
a cross-cultural perspective was completed by Dyal (1984). Anum- 
ber of alternative scales have been developed to measure control, 
including some for use with children. 

A final topic to mention in the area of "typical" performance 
is depression. Often regarded as the common cold of psychological 
disorders, this phenomenon apparently cuts across gender, age, 
culture, ethnicity, and any other independent variable that one can 
name. Unless one happens to be an android, variations in mood 
or affect are simply part of being human. A number of authorita- 
tive books dealing with the manifestations and measurement of 
depression have been written (Kleinman & Good, 1985; Mar sella, 
Hirschfeld, & Katz, 1987). And the phenomenon is, of course, an 
important topic in all theories of personality and psychopathol- 
ogy and occupies many pages in diagnostic guidelines such as 
the Diagnostic and Statistical Manual of Mental Disorders. A com- 
prehensive analysis of depression will have to give some atten- 
tion to genetic components; the assessment of neurotransmitters; 
and behavioral, cognitive, familial, and social factors, as well as 
whether and how all of these potential contributors to the condi- 
tion interact with culture. A key issue is the extent to which there 
is a "common core" set of symptoms in depression, as opposed to 
the view that the condition (if indeed it occurs everywhere) is 
manifested differently — cognitively, socially, and behaviorally — 


in different cultures. The task facing the researcher who desires to 
measure depression is potentially very complex. If a universally 
valid measure of depression existed, there would be no need for 
concern. Since such a measure is not available, the researcher will 
have to make a choice: select one of the many existing measures 
of depression or develop some ''emically sensitive'' procedure 
(Manson, 1993; Manson, Shore, & Bloom, 1985). There are many 
existing scales and techniques for measuring depression, includ- 
ing the Beck Depression Inventory, the Zung Self-Rating Depres- 
sion Scale, the Center for Epidemiologic Studies Rating Scale, the 
Depression Adjective Check List, and the depression scale (Scale 
4) of the MMPI. There are also rating scales and interview sched- 
ules, including the Hamilton Rating Scale for Depression and the 
World Health Organization's Schedule for the Standardized As- 
sessment of Depressive Disorders. All of these measures are based 
on Western criteria associated with depression, and all of them 
have been used across cultures or among ethnic /racial groups in 
the United States. This embarrassment of riches makes it a daunt- 
ing task to select which one, if any, is to be used. 

Response Sets 

Many different ways or styles of responding to scales and ques- 
tionnaires have the potential to affect data in important ways. 
These "response styles" could contribute to test score variances 
in such a way that the stylistic responses may give more informa- 
tion about the subject than the pattern of responses to the mea- 
suring device. Fiske (1970) summarized the various responses sets. 
Among the more common ones are the following: 

• Acquiescence. The tendency of subjects to agree 
("yeasayers") with just about everything or to disagree 

• Social desirability. The tendency to respond in ways that 
are more consistent with popular cultural opinions. 

In contemporary society, this response set might well be 
labeled "political correctness" — the tendency to respond 
in ways consistent with prevailing views on such things 
as the environment, foreigners, politicians, or economy. 


• Evasiveness. The tendency to avoid being forthright and 
honest in responding. 

• Carelessness. Making inconsistent, sloppy judgments. 

• Extreme response style. Obviously, consistently responding 
toward the anchor points of a scale (either 1 or 7 on a 7- 
point scale). 

• Positional response style. Responding, for instance, to first, 
last, or middle-of-the-road statements. 

The responses to any scale or questionnaire whose format 
makes it possible to respond in one or more of the above ways should 
be systematically checked for the presence of such tendencies. 

A Variant of Measures of Typical 
Response: Tine Measurement of 
Attitudes, Beliefs, and Values 

It is quite common that the researcher will be involved in projects 
that do not require the use of standardized intelligence tests or 
personality measures of the type mentioned on previous pages. 
The largest category in this domain of standardized (but not copy- 
righted or manual driven) assessment involves the measurement 
of attitudes, beliefs, and values. This section will briefly review 
some of the main considerations in this area of assessment. 

Attitude assessment has a long tradition and can be consid- 
ered to be a specialty within the broad province of social psychol- 
ogy. Many recent, comprehensive texts in social psychology at 
least contain discussions of various theories explaining how atti- 
tudes are formed and changed; some of them may also include 
information about the development and use of different attitude 
scales. And there are many books that contain detailed informa- 
tion about the construction and use of different scales that can be 
used effectively to measure attitudes. A detailed chapter summa- 
rizing the cross-cultural assessment of attitudes and beliefs ap- 
pears in Volume 6 of the Handbook of Cross-Cultural Research 
(Davidson & Thompson, 1980). 

Against this backdrop, three widely used scaling techniques 
will be summarized. 


Likert Scales 

Such scales consist of a series of statements, each followed by 
(typically) five response alternatives. A typical item might read 
as follows: 


Strongly Agree 




Strongly Disagree 

The response alternatives usually carry preassigned weights (for 
example. Strongly Agree statements may be scored 5, while 
Strongly Disagree statements may receive a score of 1). Thus, Likert 
scales can result in summated rating scales, which can result in the 
assignment of a score or scores that can be used for a number of 
purposes. Any assessment program that uses Likert scales should 
first benefit from a thorough review of the relevant literature gov- 
erning their general use. The assessment of attitudes can be com- 
plicated and controversial. 

Thursfone Scales 

Such scales are constructed so that they include equal-appearing 
intervals — that is, they are built so that the distance between any 
two adjacent points on the scale is the same. Thurstone scales use 
an 11 -point scale ranging from 1 (usually the most unfavorable or 
negative) to 11 (the most favorable or positive). Point 6 on the 
scale is neutral because it is mid-way between the ''positive'' and 
"negative" poles. After the weights have been assigned, the state- 
ments can be read by the respondent who is asked to check each 
item that approximates his or her attitude toward some object, 
idea, or institution. 

Below are three items from one of the Thurstone's Attitude 
Toward Church Scale, one of the first scales to use the technique. 
The numbers in parentheses are the scale values assigned by 
judges that are associated with the item. 



• I believe church membership is almost essential to living 
life at its best. (1.5) 

• I believe in religion, but I seldom go to church. (5.4) 

• I regard the church as a static, crystallized institution, and 
as such it is unwholesome and detrimental to society and 
the individual. (10.5) 

In preparing Thurstone scales, a group of people called 
''judges" are used to assign weights to each of the 11 positions on 
the continuum. Thus, a person's "score" would typically be the 
sum of the weights that were preassigned to each statement by 
the judges. Just as it is important to study the details of Likert 
scaling, it is important to become familiar with standard proce- 
dures for developing Thurstone scales. 

Semantic Differential Scales 

The use of the popular semantic differential (SD) technique can 
be traced to Osgood, Suci, and Tannenbaum's (1957) book, The 
Measurement of Meaning. The technique is fairly simple. Stimulus 
words or concepts (e.g., mother, peace, work) are rated on a scale 
between a series of polar opposite adjectives. Thus, a typical for- 
mat might appear as follows: 


Good : : : : : : Bad 

Tough : : : : : : Soft 

Smart : : : : : : Stupid 

The SD purports to measure metaphorical meaning and espe- 
cially affective meaning. Three dominant factors have been found 
to be used in the communication of metaphorical meaning: the 
evaluative factor (e.g., Good-Bad), the potency factor (e.g., Strong- 
Weak), and the activity factor (e.g., Fast-Slow). According to 
Osgood, these three factors account for about 60 percent of the 
variance involved in the communication of affect. And, Osgood 
argues, these factors are stable across all psycholinguistic groups 
because of the essential similarity of human languages and the 
human emotional system. 

There is considerable literature involving the use of the SD tech- 
nique with different psycholinguistic groups in many different cul- 


hires and societies. For instance, Triandis's pioneering research on 
cultural differences in roles, beliefs, and attitudes used the tech- 
nique quite effectively (Holtzman, 1980). The main source of infor- 
mation about the psychometric properties and use of the SD, with 
special attention focused on cross-cultural use of the technique, is 
Osgood, May, and Miron (1975). These two sources, especially the 
latter, should be consulted if it appears that SD technology might 
be useful in the measurement of certain attitudes and beliefs. 

The Measurement of Values 

There have been many attempts to measure human values across 
cultures. Two recent advances in such measurement should be 
consulted. The first is Hofstede's "work-related'' values. Hofstede 
(1980) claims to have found four values: uncertainty avoidance, 
power distance, individualism, and masculinity-femininity It has 
been argued that these values transcend the workplace and can 
be helpful in understanding how human values guide our actions 
in various domains of behavior. Hofstede's research was based 
upon the analysis of responses to about 117,000 questionnaires 
completed by employees of IBM, making it one of the largest data 
pools in all of psychometric history. A review of Hofstede's work 
would be beneficial to those who might want to measure the ex- 
tent to which people or groups possess the values that he claims 
to have uncovered. The value of individualism (and its polar op- 
posite, collectivism) has been researched especially heavily 

The Israeli psychologist Shalom Schwartz is currently active 
in measuring values. Schwartz (1994) has built an argument, sup- 
ported largely by a number of descriptive surveys and responses 
to his scale, that there are 10 basic human values. A review of his 
position and procedures is available (Schwartz & Bilsky, 1990), 
and he recently published a detailed account of his ideas of hu- 
man values (Schwartz, 1994). 

Nonstandardized Assessment 

All the problems associated with standardized assessment, es- 
sentially everything that has been presented so far in this chapter, 
have increased the tendency for many to seek other ways to 


assess the various domains of interest. Thus, nonstandardized, 
nontraditional, and unorthodox assessment offers alternative 
modes of assessment; one is not, after all, obligated to pick and 
choose from among the thousands of commercially available tests. 
Burrisch (1984, p. 219) offered words of encouragement in the 
domain of personality measurement when he said that ''it cost 
me two hours and a bottle of wine to write an aggression and a 
depression scale that turned out to be of equal or superior valid- 
ity, compared to much more sophisticated instruments/' 

This issue was discussed in 1981 during the 10-year reprise of 
the NATO-sponsored Istanbul meeting, mentioned earlier in this 
chapter. In that discussion, Trimble, Lonner, and Boucher (1983) 
noted that within the body of orthodox measurement theory a 
few basic assumptions stand out. These assumptions are as follows: 

• Psychometric categories that are grounded in linearly 
conceptualized mathematical frameworks do exist. It is 
assumed that this is a universal process and that every- 
one can do it. Examples are the ubiquitous scaling tech- 
niques, such as the one developed by Likert or the seven- 
point scale developed by Osgood et al. (1957) for use with 
their SD. 

• Individuals are able to generate psychosocial judgments 
about social and psychological stimuli by resorting to 
contrasting cognitive mechanisms. The processes of 
comparison, rankings, contrasts, and the like, which 
people in the Western world use with thoughtless regu- 
larity, may occur much less frequently 

• Individuals are capable of self-assessment by using 
evaluative and reflective cognitive processes. This as- 
sumption is born of the self-reflective Western world, 
where it is common to think of self as an object. Research 
concerned with relationships between self and culture 
(Markus & Kitayama, 1991) suggest wide differences in 
the individual's capacity to engage with equal facility in 
such tasks. 

While cautions about nonstandard assessment may be made, 
little can be said specifically regarding how this situation may be 


redressed or remedied. In other words, giving suggestions regard- 
ing alternatives to standardized assessment may violate a pre- 
scription that is being advanced here. Once a suggestion is made 
by someone outside who likely has an inadequate grasp of what it 
is that someone is trying to assess, there is a real risk of disrupting 
spontaneity and creativity that those inside should be encouraged 
to use. However, the following examples are presented as inter- 
esting nonstandard assessment procedures. 

The influential Swiss psychologist Jean Piaget apparently es- 
chewed standardized assessment that might give insight into the 
cognitive activities of children, in whom Piaget was most inter- 
ested. Instead, he developed his methode dinique (clinical method), 
which was essentially a detailed one-on-one dialog with a child 
during the child's attempt to solve a problem or perform a task. 
Piaget would ask a child to explain what he or she is doing and 
why. More often than not, he was interested in the errors the child 
might make rather than correct answers. After all, Piaget could 
readily grasp why a child arrived at the correct answer. What in- 
terested him more was why someone would make an incorrect 
response. The cognitions leading to an incorrect response or solu- 
tion (to an outsider) may be more elaborate and even more con- 
vincing than the processes that preceded the correct response. Note 
that this and any other "clinicar' method is deeply "ideographic,'' 
as opposed to "nomothetic." It is the individual as a genuine one- 
of-a-kind processor of information who is at the center of atten- 
tion and not a large group of individuals whose responses might 
be collated to form a unified whole in the form of group norms. 

Another example comes from attempts to understand intelli- 
gence as a cultural construction (Serpell, 1993) and from a highly 
indigenous perspective (Berry & Bennett, 1989) concerning rela- 
tionships between literacy and cognitive development. The latter 
researchers studied how Cree Indians in Canada respond when 
asked to explain what they think is associated with the idea of an 
intelligent or competent person. The method they used was rather 
simple. They used both ethnographic and psychometric proce- 
dures to uncover what the Cree understand by such notions as 
"intelligent," "smart," "clever," "able," and "competent." Work- 
ing with a small group of Cree, they collected a list of 20 words 



dealing with cognitive competence through a series of very loosely 
structured interviews conducted with Cree speakers. 

The 20 words were written out in the Cree syllabic script on 
cards. The cards were given to participants, all of whom were 
able to read the script. They were asked to put the cards into piles 
on the basis of similarity of meaning. Multidimensional scaling 
revealed two dimensions. One dimension involved negative to 
positive evaluation, with the possible inclusion of a moral dimen- 
sion as well. The other dimension was anchored on one end by 
words like "mentally tough,'' ''courage," and "fortitude." The 
other end was anchored by words like "religious" and "under- 
stands new things." Further details of the complexities of this 
example of trying to understand how indigenous people define 
intelligence or competence in their own terms can be studied in 
their original sources. 

A third example involves an attempt to develop a cognitive 
map of depression. Manson (1993) explains how he and two col- 
leagues tried to dimensionalize the depressive experience in 
American-Indian communities in contrast with other groups 
(Manson, Shore, & Bloom, 1985). Using a variation of Q-sort meth- 
odology, they asked American subjects, medical students in psy- 
chiatric residency, and American Indians to sort 100 3-by-5-inch 
cards into as many different piles as they felt necessary. On each 
card was a word or phrase frequently associated with depression 
and anxiety There were striking differences between the groups, 
both with respect to how many piles of cards were developed 
and the content of the pile. Standard depression scales may easily 
miss the patterns discovered through a technique that merely re- 
quires a little ingenuity, and a bottle of wine, to develop. 

A Special Psychometric Case: 
Projective IVIethodology 

The use of projective techniques in cross-cultural assessment has 
a long history. Many researchers — primarily ethnopsychiatrists, 
clinically oriented cultural anthropologists (or anthropologically 
oriented clinical psychologists), and certain other researchers — 
have enjoyed some success in using projectives. While a wide 


range of techniques has been employed during the past 50 or 60 
years, most who had the courage to use these techniques in other 
cultures or among different ethnic groups have used the Thematic 
Apperception Test (either in standard format or modified), the 
Rorschach, or the Holtzman Inkblot Technique. 

Many investigators were attracted to projectives because of their 
potential flexibility in the investigation of many personological 
variables. The standard definition of projectives given by Lindzey 
(1961) is admittedly inviting: 

... a projective technique is an instrument tliat is considered es- 
pecially sensitive to covert or unconscious aspects of behavior, it 
permits and encourages a wide variety of subject responses, Is 
highly multidimensional, and it evokes unusually rich or profuse 
response data with a minimum of subject awareness concern- 
ing the purpose of the test. Further, it is very often true that the 
stimulus material presented by the projective test is ambiguous, 
interpreters of the test depend upon holistic analysis, the test 
evokes fantasy responses, and there are no correct or incorrect 
responses to the test. (p. 45) 

Projectives, however, are currently used infrequently in the 
more sophisticated cross-cultural research. The main reason for 
this is based on the many pitfalls that have surfaced over the years 
of using them incorrectly, or coming up with inconclusive and 
ambiguous results. Lindzey (1961) remains the classic source guid- 
ing the use (or details of misuse) of projectives, and Spain (1972) 
reviewed their use in the field of psychological anthropology. A 
relatively recent, and key, source of information about cross- 
cultural applications of projective methodology is Holtzman's 
(1980) review. 

The researcher who might be tempted to use psychometrics 
to gather "rich and profuse'' data should keep in mind that it takes 
special training to use these techniques properly. The Rorschach 
expert, for instance, must be very familiar with quite complex 
scoring systems, most of which have not been validated satisfac- 
torily in various cultures. Moreover, one who uses projectives typi- 
cally holds rather doctrinaire viewpoints (usually psychodynamic) 
or at least believes that the Freudian principle of psychic 


determinism (there is meaning in all of our thoughts and fanta- 
sies) is alive and well. In general, this controversial area of psy- 
chometrics and culture must be approached with utmost caution. 
If the saying that caveat emptor should be applied to any area of 
psychological assessment across cultures, it is clearly in the color- 
ful domain of projectives. 

Concluding Comments 
and Summary 

Psychometrics and culture constitute a perspective that owes its 
identity to the marriage of two separate yet, in the context of this 
chapter, mutually dependent orientations. One of them is cross- 
cultural psychology, and the other is traditional psychological test- 
ing and assessment. The former is an approach that, in general, 
seeks to extend the range of variation regarding different psycho- 
logical constructs such as intelligence, depression, and values by 
seeking cultures or ethnic groups that may serve as ''test cases'' 
for a particular theory or hypothesis (see Lonner & Berry, 1986, 
and Berry et al., 1992, for details about cross-cultural methodol- 
ogy). The latter orientation is a large subdivision in the field of 
psychology. Psychometrists are concerned with technical aspects 
such as validity (Does a test measure what it purports to mea- 
sure?), reliability (Does a test measure a phenomenon or construct 
consistently?), norms (Are individuals compared with appropri- 
ate reference groups?), scoring and scaling, and even such logisti- 
cal matters as proper and fair administration of tests and scales. 
All these technical considerations have filled hundreds of volumes 
and many thousands of publications in the highly specialized 
psychometric literature. 

When cross-cultural psychology and psychometrics are joined, 
quite naturally a number of definitional and methodological prob- 
lems arise. This chapter has summarized some of the key prob- 
lems that emerge through this marriage. Essentially, what the 
conjoining of these two approaches attempts to do can be sum- 
marized by a simple sentence: the testing and assessment of indi- 
viduals in other cultures and ethnic groups requires the solution 
of a number of technical and procedural problems so that any 


results that emerge are meaningful, appropriate, practical, and 
fair. Like all marriages, if the union is compatible and the result of 
hard work and understanding, then a solid understanding of some 
aspect of individual psychological fimctioning in another cultural 
or ethnic group will be forthcoming. If, however, the marriage is 
highlighted (lowlighted?) by various failures — failures such as 
solving problems of equivalence, appropriate sampling, a clear 
understanding of what constructs and ideas mean to the various 
individuals being assessed — then one can only expect acrimony 
and unproductive results of the merger. In successful conjoinments 
of cross-cultural psychology and psychometrics, the usual con- 
cerns about validity, reliability, and norms will demand the re- 
searchers' attention. Added to these usual concerns, however, are 
always specific, culture-related problems associated with appro- 
priateness and equivalence. 

More extensive details about the problems, pitfalls, and po- 
tential solutions in this area are contained in the references given 
at the end of this chapter. Psychometrics and culture occupy the 
time and talents of quite a number of individuals around the globe. 
Those who wish to probe more deeply into the details of this area 
might want to follow the progress of the International Test Com- 
mission (ITC). The ITC recently sponsored a series of meetings 
that revolved around the adaptation and use of educational and 
psychological tests. A 13-member international panel represent- 
ing a number of international organizations is looking into these 
matters. A preliminary report was recently published (Hambleton, 
1994). This report contains guidelines covering four categories: 
context, instrument development and adaptation, administration, 
and documentation /score administration. Similar concerns have 
for years attracted the attention of the American Psychological 
Association's Board of Ethnic Minority Affairs (1989). The latter 
panel is especially concerned about the ethics of testing, while at 
the same time addressing most of the same concerns of the ITC. 

There are, obviously, many books, journal articles, and other 
documents that form the basis for a complete understanding of 
the interface between culture and psychometrics. The breadth and 
depth one wishes to probe in this area depends upon the motives 
and interests of the reader. Clearly, there are now many helpful 


guidelines that will satisfy both theoreticians and those in applied 
psychological and educational areas. 


American Psychological Association, Board of Ethnic Minority Affairs (1989, 
March 7). Draft of guidelines for psychological practice with ethnic and culturally 
diverse populations. Washington, DC: Task Force on Delivery of Services to 
Ethnic Minority Populations. 

Berry, J.W. (1979). A cultural ecology of social behavior. In L. Berkowitz (Ed.) 
Advances in experimental social psychology (Vol. 12). New York: Academic Press. 

Berry, J.W., & Bennett, J.A. (1989). Syllabic literacy and cognitive performance 
among the Cree. International Journal of Psychology, 24, 429^50. 

Berry J.W., Poortinga, Y.H., Segall, M.H., & Dasen, PR. (1992). Cross-cultural 
psychology. Cambridge, England: Cambridge University Press. 

Brislin, R.W. (Ed.) (1976). Translation: Application and research. New York: Wiley. 

Brislin, R.W. (1980). Translation and content analysis of oral and written material. 
In H.C. Triandis & J.W. Berry (Eds.), Handbook of cross-cultural psychology (Vol. 
2). Boston: Allyn & Bacon. 

Brislin, R.W. (1986). Translation and content analysis of oral and written material. 
In W.J. Lonner & J.W. Berry (Eds.), Field ynethods in cross-cultural research. 
Newbury Park, CA: Sage Publications. 

Brislin, R.W., Lonner, W.J., & Thorndike, R.M. (1973). Cross-cultural research 
methods. New York: Wiley. 

Burrisch, M. (1984). Approaches to personality inventory construction: A 
comparison of merits. American Psychologist, 39, 214-227. 

Butcher, J.N. (1984). Current developments in MMPI use: An international 
perspective. In J.N. Butcher & CD. Spielberger (Eds.), Advances in personality 
assessment (Vol. 4). Hillsdale, NJ: Erlbaum. 

Butcher, J.N. (1990). MMPI-2 in psychological treatment. New York: Oxford 
University Press. 

Butcher, J.N., & Pancheri, P. (1976). A handbook of cross-national MMPI research. 
Minneapolis: University of Minnesota Press. 

Butcher, J.N., & WilHams, C.L. (1992). Essentials of MMPI-2 and MMPI-A 
Interpretation. Minneapolis: University of Minnesota Press. 

Cole, M., Gay, J., Click, J., & Sharp, D. (1971). The cultural context of learning and 
thinking. New York: Basic Books. 

Cole, M., & Scribner, S. (1974). Culture and thought: A psychological introduction. 
New York: Wiley. 


Cronbach, L.J. (1990). Essentials of psychological testing (5th ed.). New York: Harper 

Cronbach, L.J., & Drenth, P.J.D. (Eds.) (1972). Mental tests and cultural adaptation. 
The Hague: Mouton. 

Cronbach, L.J., & Meehl, P.E. (1955). Construct validity in psychological tests. 
Psychological Bulletin, 53, 281-302. 

Dahlstrom, W.G., Lachar, D., & Dahlstrom, L.E. {19S6). MMPI patterns of American 
minorities. Minneapolis: University of Minnesota Press. 

Dana, R.H. (1993). Multicultural assessment perspectives for professional psychology. 
Boston: Allyn & Bacon. 

Davidson, A.R., & Thompson, E. (1980). Cross-cultural studies of attitudes and 
beliefs. In H.C. Triandis & R.W. Brislin (Eds.), Handbook of cross-cultural 
psychology (Vol. 6). Boston: Allyn & Bacon. 

Dyal, J. A. (1984). Cross-cultural research with the locus of control construct. In 
H. Lefcourt (Ed.), Research with the locus of control construct. Vol. 3. Extensions 
and limitations. New York: Academic Press. 

Fiske, D.W. (1970). Measuring the concepts of personality. Chicago: Aldine-Atherton. 

Gardner, H. (1983). Frames of mind: The theory of multiple intelligences. New York: 
Basic Books. 

Gough, H.G. (1987). California psychological inventory: Administrator guide. Palo 
Alto, CA: Consulting Psychologists Press. 

Graham, J.R. (1990). MMP/-2.- Assessing personality and psychopathology. New York: 
Oxford University Press. 

Greene, R.L. (1991). The MMPI-2/MMPI: An interpretive manual. Boston: Allyn 
& Bacon. 

Hambleton, R.K. (1994). Guidelines for adapting educational and psychological 
tests: A progress report. European Journal of Psychological Assessment, 10, 

Hofstede, G. (1980). Culture's consequences: International differences in work-related 
values. Newbury Park, CA: Sage Publications. 

Holtzman, W.H. (1980). Projective techniques. In H.C. Triandis & J.W. Berry (Eds.), 
Handbook of cross-cultural psychology (Vol. 2.). Boston: Allyn & Bacon. 

Irvine, S.H., & Berry, J.W. (Eds.) (1983). Human assessment and cultural factors. 
New York: Plenum. 

Irvine, S.H., & Carroll, W.K. (1980). Testing and assessment across cultures: Issues 
in methodology and theory. In H.C. Triandis & J.W. Berry (Eds.), Handbook of 
cross-cultural psychology (Vol. 2). Boston: Allyn & Bacon. 

Irwin, S.H., Schaefer G.N., & Feiden, L.P (1974). Emic and unfamiliar category 
sorting of Mars farmers and U.S. undergraduates. Journal of Cross-Cultural 
Psychology, 5, 407^23. 


Kleinman, A., & Good, B. (Eds.) (1985). Culture and depression: Studies in the 
anthropology and cross-cultural psychiatry of affect and disorder. Berkeley, CA: 
University of California Press. 

Lindzey, G. (1961). Projective techniques and cross-cultural research. New York: 
Appleton-Century -Crofts. 

Longabaugh, R. (1980). The systematic observation of behavior in naturalistic 
settings. In H.C. Triandis & J.W. Berry (Eds.), Handbook of cross-cultural 
psychology (Vol. 2). Boston: Allyn & Bacon. 

Lonner, W.J. (1990). An overview of cross-cultural testing and assessment. In 
R.W. Brislin (Ed.), Applied cross-cultural psychology. Newbury Park, CA: Sage 

Lonner, W.J., & Berry, J.W. (Eds.) (1986). Field methods in cross-cultural research. 
Newbury Park, CA: Sage Publications. 

Lonner, W.J., & Ibrahim, RA. (1989). Assessment in cross-cultural counseling. In 
P. Pedersen, J. Draguns, W Lonner, & J. Trimble (Eds.), Counseling across cultures 
(4th ed.). Newbury Park, CA: Sage Publications. 

Malpass, R.S., & Poortinga, Y.H. (1986). Strategies for design and analysis. In 
W.J. Lonner & J.W. Berry (Eds.), Fields methods in cross-cultural research. 
Newbury Park, CA: Sage Publications. 

Manson, S. (1993). Culture and depression: Discovering variations in the 
experience of illness. In WJ. Lonner & R.S. Malpass (Eds.), Reading in psychology 
and culture. Boston: Allyn & Bacon. 

Manson, S.M., Shore, J.H., & Bloom, J.D. (1985). The depressive experience in 
American Indian Communities: A challenge for psychiatric theory and 
diagnosis. In A. Kleinman & B. Good (Eds.), Culture and depression. Berkeley, 
CA: University of California Press. 

Markus H., & Kitayama, S. (1991). Culture and the self: Implications for cognition, 
emotion, and motivation. Psychological Review, 98, 224-253. 

Marsella, A.J. (1987). The measurement of depressive experience and disorder 
across cultures. In A.J. Marsella, R.M.A. Hirschfeld, & M.M. Katz (Eds.), The 
measurement of depression. New York: Guilford. 

Marsella, A.J., Hirschfeld, R.M.A., & Katz, M.M. (Eds.) (1987). The measurement 
of depression. New York: Guilford. 

Mercer, J. (1988). Death of the I.Q. paradigm: Where do we go from here? In W.J. 
Lonner & V.O. Tyler (Eds.), Cultural and ethnic factors in learning and motivation: 
Implications for education. Bellingham, WA: Western Washington University. 

Munroe, R.H., & Munroe, R.L. (1993). Behavior across cultures: Results from 
observational studies. In W.J. Lonner & R.S. Malpass (Eds.), Readings in 
psychology and culture. Boston: Allyn & Bacon. 

Neimeyer, G.J. (Ed.) (1993). Constructivist assessment: A casebook. Newbury Park, 
CA: Sage Publications. 


Osgood, C.E., May, W., & Miron, W. (1975). Cross-cultural universals of affective 
meaning. Champaign, IL: University of Illinois Press. 

Osgood, C.E., Suci, G.J., & Tannenbaum, P.H. (1957). The measurement of meaning. 
Urbana, IL: University of Illinois Press. 

Paniagua, F.A. (1994). Assessing and treating culturally diverse groups. Newbury 
Park, CA: Sage Publications. 

Poortinga, Y.H. (1983). Psychometric approaches to intergroup comparison: The 
problem of equivalence. In S.H. Irvine & J.W. Berry (Eds.), Human assessment 
and cultural factors. New York: Plenum. 

Poortinga, Y.H. (1989). Equivalence of cross-cultural data: An overview of basic 
issues. International Journal of Psychology, 24, 737-756. 

Przeworski, A., & Teune, H. (1970). The logic of comparative social inquiry. New 
York: Wiley 

Schwartz, S.H. (1994). Beyond individualism/collectivism: New cultural 
dimensions of values. In U. Kim, H.C. Triandis, C. Kagiteibasi, S. Choi, & G. 
Yoon (Eds.), Individualism and collectivism: Theory, method and applications. 
Newbury Park, CA: Sage Publications. 

Schwartz, S.H., & Bilsky, W. (1990). Toward a theory of the universal content 
and structure of values: Extensions and cross-cultural replications. Journal of 
Personality and Social Psychology, 58, 878-891. 

Sechrest, L., Fay, TL., & Zaidi, S.M.H. (1972). Problems of translation in cross- 
cultural research. Journal of Cross-Cultural Psychology, 3, 41-56. 

Segall, M.H., Dasen, PR., Berry J.W., & Poortinga, Y.H. (1990). Human behavior in 
global perspective: An introduction to cross-cultural psychology. New York: 

Serpell, R. (1993). The cultural construction of intelligence. In W.J. Lonner & R.S. 
Malpass (Eds.), Readings in psychology and culture. Boston: Allyn & Bacon. 

Sharma, S. (1977). Cross-cultural comparisons of anxiety: Methodological 
problems. Topics in Culture Learning, 5, 166-173. 

Spain, D.H. (1972). On the use of projective techniques for psychological 
anthropology. In F.K. Hsu (Ed.), Psychological anthropology. Cambridge, MA: 

Spearman, C.E. (1904). General intelligence objectively determined and 
measured. American Journal of Psychology, 15, 201-293. 

Spielberger, C. (1984). A comprehensive bibliography for the State-Trait Anxiety 
Inventory. Palo Alto, CA: Consulting Psychologist Press. 

Sternberg, R.J. (1985). Beyond IQ: A triarchic theory of human intelligence. New 
York: Cambridge University Press. 

Sundberg, N.D. (1977). The assessment of persons. New York: Prentice-Hall. 


Thorndike, R.M. (1988). Commentary: Science and the IQ paradigm. In W.J. 
Lonner & V.O. Tyler (Eds.), Cultural and ethnic factors in learning and motivation: 
Implications for education. Bellingham, WA: Western Washington University. 

Triandis, H.C. (Ed.) (1980-1981). Handbook of cross-cultural psychology. Boston: 
AUyn & Bacon. 

Triandis, H.C, & Berry, J.W (Eds.) (1981). Handbook of cross-cultural psychology 
(Vol. 2). Boston: Allyn & Bacon. 

Triandis, H.C, & Brislin, R.W. (1984). Cross-cultural psychology. American 

Trimble, J.E., Lonner, W.J., & Boucher, J.D. (1983). Stalking the wily emic: 
Alternatives to cross-cultural measurement. In S.H. Irvine & J.W Berry (Eds.), 
Human assessment and cultural factors. New York: Plenum. 

Vernon, RE. (1969). Intelligence arid cultural environment. London: Methuen. 

Whiting, B., & Whiting, J. (1975). Children of six cidtures: A psychocultural analysis. 
Cambridge, MA: Harvard University Press. 

Witkin, H.C, & Berry, J.W. (1975). Psychological differentiation in cross-cultural 
perspective. Journal of Cross-Cultural Psychology, 6, 4-87. 

Witkin, H.C, & Goodenough, D. (1981). Cognitive styles: Essence and origins. 
In Psychological monographs, 51. New York: International Universities Press. 

Wober, M. (1966). Sensotypes. Journal of Social Psychology, 70, 181-189. 



The Role of Ethics in 

Evaluation Practice: 

Implications for a 

Multiethnoculturol Setting 

Wei Li Fang, Pli.D. 

Ethics tielps us to understand ourselves as responsible beings, 
our world as the place in which the responsible existence of the 
human community is exercised. Its practical utility is in its clarifi- 
cation, its interpretation, its provision of a pattern of meaning 
and understanding in the light of which human action can be 
more responsible.... It does not relieve [people] of personal re- 
sponsibility to exercise their freedom, their capacity for judg- 
ment and action in the world. But each man is not left to himself 
alone, though each is personally responsible. (Gustafson, quoted 
in Smith, 1985b, p. 9) 

In recent years, evaluators have increasingly turned their at- 
tention to ethics and its implications for guiding the development 
of evaluation designs and the implementation and use of evalua- 
tion findings. The purpose of this chapter is to review briefly the 
historical origin of an ethics model, followed by a discussion of 
moral issues as they relate to evaluation practice. When ethics is 
applied to multiethnocultural settings, there are additional dimen- 
sions to be considered. These dimensions and potential ethical 
dilemmas or pitfalls are then discussed in the context of an 


evaluation study. An ethical evaluation model is then offered as a 
possible strategy. 

Biomedical iVIodei of Etinics 

As evaluators, our view of ethics is based largely on a traditional 
biomedical model of ethics, in which moral reasoning can be 
viewed in the form of a hierarchy, where ''judgments about what 
ought to be done in particular situations are justified by moral 
rules, which in turn are justified by principles, which ultimately 
are justified by ethical theories'' (Beauchamp & Childress, 1983, 
p. 5).^ This moral reasoning is applied to the medical setting in 
order to protect patients from potential abuse by medical research- 
ers (Thorne, 1980). According to this model, the patient is viewed 
as the subject, and the investigator controls what the subject is to 
do. Investigators are expected to provide full disclosure about the 
study; this includes potential risks and benefits of the study to 
both the patient and society, as well as the patient's rights to health 
care, whether he or she decides to participate or withdraw. Confi- 
dentiality of patient data must also be ensured. If a patient de- 
cides to participate, he or she must then grant written informed 

Although some of these elements of the biomedical ethics 
model are relevant to evaluation, important differences exist in 
the investigator-participant relationship and in the procedures by 
which the studies are conducted. Whereas the investigator- 
participant relationship can be considered asymmetrical, with the 
investigator wielding the power in a biomedical experiment 
(Cassell, 1982), an evaluation study ideally strives for a more col- 
laborative approach. In an evaluation study, subjects are known 
as participants or informants, while investigators are evaluators 

'A judgment expresses a decision, verdict, or conclusion about a par- 
ticular action.... [R]ules state that actions of a certain kind ought (or 
ought not) to be done because they are right (or wrong).... Principles 
are more general and fundamental than moral rules and serve as 
their foundation or source of justification.... [TJheories are bodies of 
principle and rules, more or less systematically related" (Beauchamp 
& Childress, 1983, p. 5). 


or field workers. Ethical dilemmas may arise because of the more 
loosely defined boundary between evaluator and informant and 
the qualitative nature of the study itself (Ramos, 1989). 

Moral Issues 
Evaluation Moral Issues 

Evaluation moral issues (e.g., professional standards of ethical con- 
duct) are distinct homprogram moral issues (i.e., the extent to which 
programs are fulfilling their objectives and the impact that these 
programs have on the lives of those they serve) (Worthen & Sand- 
ers, 1987). The profession has focused more attention on evalua- 
tion moral issues, or the practice of evaluation; this is indeed 
evident in the standards for evaluations of educational programs, 
projects, and materials that were issued by the Joint Committee 
on Standards for Educational Evaluation in 1981 and reorganized 
and reissued in 1994. This professional code of ethics is rooted in 
the biomedical model of ethics, which requires individuals to re- 
flect on the moral principles of beneficence, respect, and justice in 
order to make ethical decisions. 

• Beneficence. Maximizing good outcomes for science, 
humanity, and the individual research participants while 
avoiding or minimizing unnecessary risk, harm, or 

• Respect. Protecting the autonomy of (autonomous) per- 
sons, with courtesy and respect for individuals as per- 
sons, including those who are not autonomous (e.g., 
infants, the mentally retarded, senile persons). 

• Justice. Ensuring reasonable, nonexploitative, and care- 
fully considered procedures and their fair administration; 
fair distribution of costs and benefits among persons and 
groups (i.e., those who bear the risks of research should 
be those who benefit from it) (Sieber, 1992, p. 18). 

These moral principles are embodied in the propriety stan- 
dards that "are intended to ensure that an evaluation will be con- 
ducted legally, ethically, and with due regard for the welfare of 


those involved in the evaluation, as well as those affected by its 
results'' (Joint Committee on Standards for Educational Evalua- 
tion, 1994, p. 81). Propriety standards include service orientation, 
formal agreements, rights of human subjects, human interactions, 
complete and fair assessment, disclosure of findings, conflict of 
interest, and fiscal responsibility. 

This code of professional ethics thus delineates the expecta- 
tions of evaluators in relation to ethical problems that they may 
encounter. Because every, single instance cannot be defined, the 
standards can only serve as a useful guideline and aid to novice 
evaluators about potential pitfalls. Professional ethics are distinct 
from personal ethics in that 

being ethical is a broad, evolving personal process that both 
resembles and is related to the process involved in becoming a 
competent social scientist. Ethical problems in program evalua- 
tion are problems having to do with unanticipated conflicts of 
obligation and interest and with unintended harmful side ef- 
fects of evaluation. To be ethical is to evolve an ability to antici- 
pate and circumvent such problems. It is an acquired ability.... 
As one undertakes new and different kinds of evaluation and as 
society changes, one's ability to be ethical must grow to meet 
new challenges. Thus, being ethical in program evaluation is a 
process of growth in understanding, perception, and creative 
problem-solving ability that respects the interests of individuals 
and of society. (Sieber, 1980, p. 53) 

Program Moral Issues 

Even though evaluation moral issues are routinely addressed by 
evaluators, program moral issues are often neglected in spite of 
widespread agreement that the role of the evaluator is to deter- 
mine the value or worth of a social program or practice. Indeed, 
responsibility for assessing the value of important social programs 
and policies is assumed by evaluators, and their findings and rec- 
ommendations frequently play a decisive role about a program's 
future (Jerrell & Jerrell, 1985; Shadish, Cook, & Leviton, 1991). 
According to Ericson (1990, p. 6), there are three reasons for this 



First, there is the political context of evaluation practice. Sec- 
ond, there is the view that the practice of evaluation extends 
only to cover social programs, policies, and practices, not insti- 
tutions (thetme subject of social justice) in which the former are 
merely embedded. And third, despite the intent to determine 
the value and worth of a program, evaluators generally disclaim 
any responsibility forjudging the ends or aims of the programs. 

An analysis of program moral issues calls for an analysis of 
the moral principles on which the program is based. According to 
some evaluators (Ericson, 1990; House, 1990; Schwandt, 1992; 
Smith, 1985b; Stake, 1986), evaluation practice must address the 
core principles and values of individuals and institutions or fall 
short of its mandate to probe into the important role that the con- 
cept of social justice has to play in the evaluation of social pro- 
grams, practices, policies, and institutions. House (1990, pp. 23-24) 

Social justice is among the most important values we should 
hope to secure in evaluation studies. The contemporary practice 
of evaluation is part of the political authority structure of soci- 
ety, and evaluation as an aid to public decision making entails 
conceptions of democracy and social justice, even when these 
conceptions are not immediately apparent. Public evaluation 
should be an institution for democratizing public decision mak- 
ing, for making decisions, programs, and policies more open to 
public scrutiny and deliberation. As an institutionalized practice, 
it should conform to the values and ethics of a democratic soci- 
ety. Considerations such as justice, impartiality, and equality, while 
subject to disagreement and debate on their exact meaning, are 
neither arbitrary or relative. 

Evaluations should serve the interests not only of the sponsor 
but of the larger society and of various groups within society, 
particularly those most affected by the program under review. 
Hence, as a social practice, evaluation entails an inescapable ethic 
of public responsibility responsibility that extends well beyond 
the immediate client. Social justice in evaluation, then, concerns 
the manner in which various interests are served. 


Ericson (1990) concluded that "an evaluation that simply at- 
tempts to identify in a summative or formative manner actual or 
likely program consequences is necessarily incomplete ... evalua- 
tion practice cannot escape the necessity of raising and dealing 
with the normative and ethical concerns naturally embedded 
within educational practice'' (p. 19). However, he noted that evalu- 
ators often shirk this responsibility for a variety of reasons: 

[Ejvaluators may feel that they risk (1) alienating many of their 
various audiences, (2) adding to the explosiveness of a situation, 
(3) undermining the utility of the evaluation, and (4) undermin- 
ing their own credibility as objective arbiters. ...In adhering to 
evaluation standards that prescribe evaluator sensitivity to the 
"needs" of various constituencies and to the actual distribution 
of program benefits and burdens, evaluators are often in the 
position to judge when the requirements of justice are ignored. 
To ignore inequities demeans both the virtues of courage and 
integrity that is a key part of evaluation demands. (Ericson, 1990, 
pp. 6-7) 

Implications for Multiethnoculturol 

The recent emphasis on multicultural diversity has its own impli- 
cations for ethics. Evaluators must be vigilant about the develop- 
ment and implementation of evaluation designs and instruments 
that are culturally sensitive and meet the needs of the various 
stakeholders. The understanding of cultural norms, values, and 
behavioral codes will enhance an evaluator 's ability to collect, 
analyze, and disseminate accurate and useful information. Madi- 
son (1992) advocated the primary inclusion approach. According 
to this approach, program participants who represent a variety of 
racial and ethnic groups are encouraged to become involved in 
all phases of program development, from the conceptualization 
of problems to the evaluation and the interpretation of findings. 
The inclusion of the various stakeholders may require actively 
seeking their opinions, in their own settings and on their own 


terms, in order to ensure that relevant ethnic and racial groups 
are represented (Orlandi, 1992a). 

The evaluator needs to see, as much as is reasonably possible, 
the problems from the perspective of the targeted group. Failure 
to take cultural dimensions into reasonable consideration can lead 
to misinterpretations of social reality or various perceptions 
thereof. According to Madison, "Definitions of social problems 
must emerge from constructions of reality within the cultural con- 
texts and experiences of the individuals affected by the solutions. 
If the cultural context is not incorporated into our understanding 
of the environments where we define problems and conduct evalu- 
ations, then evaluation may be limited in its ability to inform'' 
(1992, p. 37). This sentiment is echoed by Davis (1992, p. 60): 

The contemporary practice of program evaluation occurs within 
social units and environments; hence, evaluations are likely to 
be circumscribed and often constrained by social and political 
structures. With this in mind, we can see that program evalua- 
tion becomes a social practice that influences how evaluators 
construct the social realities of program participants and how 
they analyze results. Their analyses are affected by their under- 
standing of participants' social and cultural experiences. . . . Evalu- 
ations of programs for culturally diverse populations should not 
be based on assumptions inherent in a positivist evaluation frame- 
work; evaluation should be more sensitive to the nature of the 
social construction of the programs. 

The primary inclusion approach takes time, which may present 
a challenge to evaluators who work under strict time constraints. 
The process may often be intimidating because evaluators may 
not be accustomed to sharing power and working cooperatively 
with individuals who differ in terms of education, class, culture, 
or racial or ethnic group from themselves. Evaluators come with 
their own biases and may not have the sensitivity or experience 
to work with people of various racial, ethnic, or socioeconomic 
backgrounds. They may not be familiar with their clients' cultural 
norms and customs. However, the primary benefit of using this 
approach is the integration of the targeted group's perspective 

into the conceptual definition of the problem, thus leading to an 
increase in the chances of the program's success in meeting the 
participants' needs (Grace, 1992; Madison, 1992; Stockdill, Duhon- 
Sells, Olson, & Patton, 1992). 

In the end, the primary inclusion approach may save time, in 
that the various stakeholders become involved in the evaluation 
process, develop ownership, and have an investment in seeing 
the program succeed. Madison (1992, p. 40) provides the ratio- 
nale for including the perspectives of various cultures and racial 
and ethnic groups in this process: 

First, input from tlie intended beneficiaries of programs may 
enlighten evaluators and program developers about social reali- 
ties as they are experienced. Second, the program beneficiaries 
are in a better position than anyone else to explain what has 
worked and what has not worked for them in the past, allowing 
for better understanding of cause-effect relationships between 
program interventions and outcomes. Such information could 
lead to more efficient use of resources. Third, program partici- 
pants may provide an opportunity for observation of positive 
models, rather than the deficit models usually used in the con- 
struction of social theories about minorities. 

Madison (1992) promoted integrative approaches to program 
theory development because they incorporate both stakeholders' 
viewpoints and social science knowledge. Thus, program staff, 
intended program beneficiaries, and other stakeholders are in- 
cluded in the knowledge-generating process and collaborate on 
ensuring that the cultural context is integrated into descriptions 
about the program and explanations about its effectiveness. It also 
builds in member checks so that disparities in cultural interpreta- 
tions of what is occurring in the program are discussed and 
reconciled. Patton (1985, p. 94) coined the term ''situational re- 
sponsiveness," which includes 

sensitivity to culture in all its manifestations: political culture, 
program culture, organizational culture, local community cul- 
ture, interpersonal norms, societal traditions, and local cultural 
values. Situational responsiveness begins with the recognition 


that evaluation is a culturally bound activity. Evaluation research 
is a subcultural perspective and set of practices within the larger 
culture of science. Program staff and clients will vary in the extent 
to which they understand and share the empirical orientation 
that is central to scientific cultures and subcultures. Given this 
variation, every evaluation becomes a cross-cultural encounter— 
a blending or confrontation between culturally different perspec- 
tives represented by evaluators and program staff or participants. 

Charges have been made that evaluation practice has failed 
to take into account the perspectives of various cultures and ra- 
cial and ethnic groups. These groups have complained that evalu- 
ators who were neither familiar with nor sensitive to the cultural 
dynamics and characteristics of their group were employed. The 
evaluation design and strategies have not always meshed with 
the realities of the program setting and the activities that were 
targeted to and provided for specific groups. Program participants 
and evaluators have differed in their definition of what consti- 
tuted success (Stake, 1986). In addition, inappropriate standards 
on which to interpret the program and make value judgments 
were identified and used. Instruments that were not capable of 
accurately assessing the value of the program were also adminis- 
tered (Stockdill et al, 1992). 

Ethical Dilemmas Encountered in 
Evaluation Practice 

As noted earlier, evaluation is conducted in order to provide in- 
formation for decision makers. Difficulties arise when the evalu- 
ator is in conflict with the client on ethical grounds (Adams, 1985; 
Mathison, 1991; Morris & Cohn, 1992; Newman & Brown, 1992; 
Pope & Vetter, 1992; Shadish & Epstein, 1987; Sheinfeld & Lord, 
1981; Sieber & Sanders, 1978). Ethical problems can, and do, how- 
ever, arise in every stage of the evaluation process (Morris & Cohn, 
1992). In the paragraphs that follow, ethical issues are presented 
by these various stages: developing the evaluation contract; de- 
signing the evaluation; collecting, storing, and analyzing data; and 
interpreting and reporting evaluation findings. In addition, the 
role of the e valuator and training issues are discussed. 


Developing the Evaluation Contract 

The first step for any evaluator is deciding whether to accept an 
evaluation contract. This decision is dependent in part on know- 
ing how one's own personal values relate to the program under 
study, the goal of the program and of the evaluation, and the types 
of program evaluation that one is willing to conduct (Covert, 
Voorhees, & Honea, 1988). A reasonable exploration of the values 
of the various stakeholders, particularly in multiethnocultural 
settings, is equally important in determining the fit of the evalua- 
tor to the program. At this stage, it is also important to determine 
whether the stakeholders are willing to negotiate and address ethi- 
cal concerns as they arise. Five ethical problems that may occur in 
contracting with stakeholders have been identified: 

(1) the stakeholder has already decided what the findings should 
be or plans to use the findings in an ethically questionable fash- 
ion; (2) the stakeholder declares certain research questions 
off-limits in the evaluation despite their substantive relevance; 
(3) legitimate stakeholders are omitted from the planning pro- 
cess; (4) various stakeholders have conflicting expectations, pur- 
poses, or desires for the evaluation; and (5) the evaluator has 
difficulty identifying the key stakeholders. (Morris 3 Cohn, 1992, 
Table 8 on p. 12) 

These challenges are helpful in alerting evaluators to poten- 
tial pitfalls that they should consider prior to program entry. It is 
useful to explore these issues when developing an evaluation con- 
tract or agreement. This mutually negotiated contract specifies 
behavioral expectations of the evaluator and the client, as well as 
products and timelines. As such, it recognizes a mutual respect 
and shared responsibility of both stakeholders and evaluators in 
conducting an ethical evaluation. This contract also considers the 
same ethical concerns that are addressed in an evaluation. Each 
of these is briefly discussed below. 

• Confidentiality of participant information. Issues of access, 
ownership, and use of the data should be specified. Of 
utmost importance is a consideration of the informants and 
a safeguarding of individuals' rights, interests, and sensi- 


tivities (Beauchamp & Childress, 1983; Diener & Crandall, 
1978; Kimmel, 1988; Patton, 1990; Spradley, 1979). The 
disclosure agreement should address the participants' right 
to voluntary informed consent and ensurance of their 
privacy and the confidentiality of the data they provide. It 
should also present the evaluator's right of access to data, 
rights of independent publication of findings, restrictions on 
release of data, the rights of others to seek independent 
reanalysis of the data, obligations to provide data to others, 
and the rights of the evaluator to participate in project 
planning and implementation that will affect the kinds of 
data that will be available (Patton, 1990; Sieber, 1992; Sieber 
& Sanders, 1978). 

Risks. The risks of collecting, entering, storing, and dissemi- 
nating evaluation findings are delineated. These include 
risks to the privacy of individuals connected with the 
project, risks to program continuation, and risks to the 
integrity or reputation of project personnel, participants, the 
funding agency, or the evaluator (Sieber & Sanders, 1978). 
Steps that will be taken to mitigate these risks are also 

Benefits. The benefits of the proposed program that are ex- 
pected by the various stakeholders are listed. Included also are 
benefits of conducting the evaluation as proposed and the 
steps to be taken to maximize these benefits (Sieber & Sanders, 

Availability of the stakeholders. Evaluators should have reason- 
able access to the various stakeholders to ensure that their 
perspectives are obtained and adequately addressed in the 
evaluation plan. Access to decision makers is particularly 
crucial in clarifying organizational policies and procedures, 
the history of the organization and its political culture, and 
the decision-making process. Introducing evaluators to the 
various stakeholders, coordinating meetings and other 
functions, and making copies of program documents and 
manuals available for review are crucial. 
Modifying the evaluation plan. A mechanism should be 
in place for modifying the evaluation plan and its 

implementation if new concerns arise on the part of any of 
the stakeholders or evaluators (Sieber & Sanders, 1978). 
Reporting and discussing evaluation findings with decision 
makers should also occur periodically, both formally and 
informally, so that stakeholders are kept informed of 
progress, potential pitfalls and problem areas, and suc- 
cesses. Evaluation is a shared activity of stakeholders and the 
evaluator and is a two-way process aimed at the mutual 
understanding of the program under study 

• Implications of withdrawal The contract should specify the 
conditions under which the evaluator or the client may 
withdraw from the agreement. These may include condi- 
tions under which the evaluator may resign, conditions 
under which the client or evaluator may make a modifica- 
tion of the evaluation agreement, conditions under which 
the rights of the targeted group must be defended although 
they are in conflict with the aims of the evaluation, and 
conditions under which the project might be terminated 
(Sieber & Sanders, 1978). 

• Fiscal integrity. A budget should specify not only the costs 
associated with the evaluation, but also a justification for 
each item. The delineation of expected products and ser- 
vices should also include a timeline. In addition, the fiscal 
integrity of accounting procedures and reporting must be 
ensured (Worthen & Sanders, 1987). Budget issues may 
cause ethical dilemmas if it is clear that an insufficient 
amount of personnel, effort, money, or resources have been 
allocated to the evaluation, yet there are limited funds 

During the development of the contract and throughout the 
course of the evaluation, the evaluator should be sensitive to the 
client's agenda and concerns by responding honestly and frankly 
to his or her perceptions, thoughts, and ideas about the program 
and its participants. Inherent in this relationship are a mutual re- 
spect and trust. The relationship between the evaluator and pro- 
gram stakeholders 


has to be built on the basis of mutual exchange, the preserva- 
tion of hunnan dignity, privacy and confidentiality, and joint ne- 
gotiation of research purposes, strategies and interpretations. 
This means nothing less than the form of inquiry which is in- 
creasingly termed collaborative or joint inquiry, wherein the re- 
searched become co-equal partners in the research effort, and 
where they have equal say in the interpretation and distribution 
of results of the inquiry The power of agency and the locus of 
control never leave the province of the researched, and their 
decisions regarding the information about them (and the real or 
possible harm that such information might bring to them) re- 
main theirs to negotiate in the present and the future. (Lincoln 
& Cuba, 1986, p. 26) 

Designing the Evaluation 

The design stage of an evaluation calls for an ethically sound and 
culturally sensitive political process whereby the various stake- 
holders are involved and equally represented. During this stage, 
stakeholders work with evaluators to identify evaluation ques- 
tions about program activities and recipients; to select informants; 
to assign evaluation responsibilities; to choose strategies and mea- 
sures; to determine field testing sites; to develop the data collec- 
tion plan; and to delineate the timeline. It is also important to 
determine the potential impact and the social justice of the pro- 
gram as well as the moral justification for designing the program 
in a certain way (Bunda, 1985). According to Madison (1992, p. 41), 

[i]n the selection of evaluation methods, the evaluator should 
consider the ability of the evaluation design to answer questions 
about social justice in terms of the value of the program in the 
context of the program participants' expectations and social re- 
alities. To address issues of social justice, a broad range of ap- 
proaches to evaluation is necessary Primary inclusion of program 
participants in the selection of appropriate evaluation strategies 
can provide cultural credibility to the interpretation and applica- 
tion of evaluation findings. Program participants can also offer 
insight into whether moral questions of social justice have been 
addressed in the evaluation design. 


Some researchers have argued that traditional approaches to 
evaluation focus on quantitative data and statistical analysis. They 
have charged that these types of evaluations have resulted in de- 
signs that hold little meaning for decision makers and other stake- 
holders and that a dichotomy between the evaluator and those 
involved in the program is established, wherein tensions arise 
due to misconceptions about evaluation and its purposes. Grace 
(1992, p. 60) contended that these approaches are frequently det- 
rimental to the evaluation process, particularly in racial and eth- 
nic communities: 

In general, traditionally trained evaluators, like many other re- 
searchers, place a great deal of importance on the objective, 
logical, rational aspects of phenomena. Consequently they tend 
to mistrust the emotional and nonquantifiable and to value highly 
the printed word. These factors, which reflect the particular cul- 
tural frame of reference being brought to the task, are signifi- 
cant in determining what is measured, what methods are used, 
and how the results are interpreted. Similarly the dominant cul- 
ture stresses such values as independence and competition and 
promotes a preference for the objective over the subjective in 
scientific inquiry These and other cultural values play a signifi- 
cant role in the selection of program goals, the approaches to 
evaluation, and the identification of variables to be examined. 
However, these values often clash with those promoted in Afri- 
can-American communities. In general, African-American cul- 
ture encourages interdependence and cooperation and finds 
more merit in subjective experience, values, and behavioral codes 
that have been linked to the ability of Black families to survive in 
the face of adversity. 

The need to adopt less traditional models of evaluation was 
reinforced by Stockdill (1992), who recoimted how her experience 
in evaluating a multicultural education program caused a shift in 
her own thinking and evaluation practice: 

/ have been forced to break away from some basic evaluation 
assumptions: that numbers are important, that giving negative 
feedback is an important function, and that evaluation principles 
can be applied to any setting. I learned that not all cultures value 


counting exercises, that tlie style in wliicii many Americans give 
negative feedbacl< directly violates the norms of some cultures, 
and that evaluation is greatly affected by the extent to which 
program and administrative decision making is democratic and 
egalitarian rather than autocratic and hierarchical, (p. 28) 

Thus, evaluation of programs in racial and ethnic communi- 
ties calls for culturally relevant and sensitive evaluation models 
such as those described and espoused by Patton (1992), Stake (in 
Shadish, Cook, & Leviton, 1991, pp. 270-314), Linney and 
Wandersman (1991), and Payne (1994). This cultural sensitivity 
means that the integrity of the racial and ethnic groups remains 
intact and that values, beliefs, attitudes, practices, and traditions 
of the program recipients need to be considered during the de- 
sign phase (Davis, 1992; Madison, 1992; Orlandi, 1992b; Sheinfeld 
& Lord, 1981). Different stakeholders frequently have disparate 
experiences with, beliefs about, and trust in ''data,'' research, and 
accountability, and these perspectives should be integrated into 
the evaluation process (Stockdill et al., 1992). Active involvement 
of program participants during the design phase is desirable so 
that they too may have a say in determining the scope of the study 
and determining potential outcomes and impact. This attention 
to the mores and culture of targeted groups and the sharing of 
power applies not only to the program itself, but also in the treat- 
ment of program recipients. Inherent in this process is the atti- 
tude of the evaluator toward program recipients and the racial 
and ethnic orientation of the evaluator. 

In terms of the program itself, past experience has indicated 
that social programs have been developed for particular groups 
or organizations and are later adopted and implemented for un- 
intended groups. This can lead to a mismatch of a program and a 
targeted group, as well as disappointing or erroneous findings 
(Davis, 1992). The selection of the unit of study (e.g., the commu- 
nity, a particular group of people, individuals) will also affect how 
the evaluation will be conceptualized. The evaluation questions 
will determine whether qualitative and /or quantitative measures 
are used, as well as the evaluation design (e.g., case study, experi- 
mental design). 

In addition, recipients of program benefits are almost always 
in a less powerful position than program providers and evalua- 
tors. This imbalance may lead to injustice even when recipients 
are those who are not generally thought of as being disadvan- 
taged. However, this injustice problem can be compounded when 
the recipient is a member of a racial, ethnic, or cultural group 
(House, 1990). 

Another consideration is the race and ethnic perceptions that 
evaluators bring to the evaluation setting. These perceptions may 
be misinformed, pejorative, and potentially harmful and may in- 
fluence the design of the study as well as the interpretation of 
results. Davis (1992) stated that evaluators must challenge their 
own biases and assumptions about the program recipients and 
observe how these perceptions can affect their decisions during 
the evaluation process. Therefore, 

decisions about the cultural appropriateness of an evaluation 
strategy must be coupled with an acknowledgment of the cul- 
tural integrity of such groups as African Americans, whose abil- 
ity to contribute to the development and evaluation of programs 
in their communities must be recognized . . . evaluators must cul- 
tivate a multiethnocultural perspective, by which they can rec- 
ognize and incorporate the experiential differences between and 
among individuals and groups. (Davis, 1992, p. 61) 

When drawing up the evaluation plan, evaluators are re- 
minded to refer to the propriety standards delineated by the Joint 
Committee on Standards for Educational Evaluation (1994). Or- 
ganizational policies and protocols must also be followed through- 
out the evaluation process (Worthen & Sanders, 1987). Certain 
ethical and legal considerations must also be addressed by the 
evaluation plan to ensure that the rights of participants are pro- 
tected. Many organizations have an institutional review board that 
assesses such factors as informed consent, privacy of the indi- 
vidual, confidentiality of participant data, identification of risks 
and benefits to the participant, alternatives to the proposed pro- 
gram, availability of the researcher to answer questions from the 
stakeholders, and implications of withdrawal once the study has 
been initiated. This review generally occurs prior to the adminis- 


tration of a survey, a project's implementation, or submission to a 
funding agency (Fang & Ellwein, 1990). 

Collecting. Storing, and Analyzing Data 

Of utmost importance in the collection, storage, and analysis of 
program data is the respect for individual and /or group autonomy 
and privacy This has particular resonance when working with 
ethnic and racial groups. Evaluators must safeguard the anonym- 
ity of their informants through the use of and adherence to stan- 
dardized protocols, numbering systems, and in some instances, 
locked storage. 

Lack of information about the cultural characteristics of the 
targeted group and lack of appropriate program goals can con- 
tribute to the failure to choose appropriate data collection instru- 
ments with which to measure program effectiveness. Many 
assessment tools were normed on groups with characteristics that 
differ significantly from the program recipients or group under 
study Field testing of these instruments may be necessary to de- 
termine their appropriateness and utility (Grace, 1992). 

The issue of self-disclosure during data collection and report- 
ing of formative evaluation data is important because it gives 
stakeholders an idea of what the evaluator is thinking. It is also a 
method for building rapport with the informant and provides the 
opportunity for stakeholders and evaluators alike to share insights 
and perceptions about the program. These self-disclosures also 
give the evaluator an opportunity to confirm hypotheses and test 
emerging conclusions with stakeholders (Patton, 1990; Taylor, 
1987; Torres, 1991). 

The presence of the evaluator, or the fact that an evaluation is 
taking place, can also affect the findings of a study in a number of 
ways. Program decision makers and staff members may react to 
the evaluation positively or negatively due to past experience or 
their perceptions of what an evaluation should be. They may have 
certain expectations or misconceptions about the evaluation it- 
self — its goal, design, implementation, or findings — and these may 
be in conflict with those of the evaluator. For example, formative 
and summative evaluation are frequently confused, particularly 


when stakeholders have little working knowledge of why and 
how they differ in intent and result. Misunderstanding often oc- 
curs when stakeholders think a process evaluation will tell them 
about the outcome or impact of the program on its recipients. 
Evaluations can also be perceived as threatening even when staff 
members receive orientation or training about the purpose and 
process of evaluation. 

Changes in the evaluator (e.g., biases, views toward the pro- 
gram staff or recipients, differential adherence to evaluation pro- 
tocols or administration of data collection instruments) during 
the course of the evaluation may also affect the findings. This is 
particularly true when qualitative methodology is used since the 
measuring instrument is the evaluator. However, instrumentation 
effects may occur daily — from stress, fatigue, illness, and other 
personal factors. Other factors that may affect the evaluation in- 
clude the predispositions or biases of the evaluator and evaluator 
incompetence (including lack of sufficient training or preparation). 
Information that is obtained in an interview may also be so dis- 
quieting or confusing that the evaluator may feel the need to seek 
debriefing and advice from a colleague or confidante (Patton, 1990; 
Stake, 1986). 

Evaluators are reminded to follow the established protocols 
as outlined in the evaluation contract and the evaluation design. 
Of equal importance is the ensurance that accepted procedures 
(e.g., informed consent, confidentiality of data, due process) are 
maintained during the collection of program-related information. 
While the purpose of an interview is to gather data, it may also 
cause psychological stress, in that the informant will be asked to 
reflect on an experience or a program or his or her life (Patton, 
1990). The interview may reveal information that may jeopardize 
continued participation in the program or rejection by program 
staff members or other program recipients. If information of an 
unethical, illegal, or troubling nature is obtained during an inter- 
view, it is important for evaluators to have contingency plans to 
work further with these individuals and to refer them to appro- 
priate professional resources. When negative information is un- 
covered, program decision makers and evaluators must have 
potential solutions or alternatives available to address these prob- 


lem areas. Social scientists should also be reminded that they 
do not have the same legal protection of clergy and lawyers 
and can be summoned to testify in court; thus, interviews can 
put the interviewee at risk. There may also be political reper- 
cussions that occur as a result of the interview. 

Ethical problems may also arise when participants are ei- 
ther photographed or videotaped since identities are clearly 
revealed. Therefore, participants must be given the opportu- 
nity to view the products that will be used in the evaluation 
report or presentation. This relates to the concept of a covenan- 
tal relationship, where the evaluator and informant make con- 
sensual decisions about the framing of the photograph, the final 
print, and its use and interpretation (Gold, 1989). Selective ed- 
iting in the case of videotapes may also present ethical prob- 
lems and thus require informed consent regarding use. 

According to Gold (1989), the use of visual data can lead to 
ethical dilemmas related to confidentiality, informed consent, 
reactions of the informants, the accompanying narrative text, 
and the conceptual control of the fieldworker. Perhaps these 
dilemmas appear daunting, but the added visual dimension 
enriches the audience's view of the program being evaluated. 
Photographs and videotapes also capture the flavor and subtle- 
ties of interaction between people and activities, as well as serv- 
ing as a method of triangulating other sources of data (Fang, 
1986). A caveat is that they must be an accurate reflection of the 
context in which they were taken. Of primary ethical concern 
are issues of honesty and accurate representation as they relate 
to sampling the context of the program being evaluated; ex- 
plaining to the informants the proposed use of the photographs 
in the evaluation; and selecting and interpreting the photo- 
graphs for the oral or written presentation (Fang & Ellwein, 1990). 

Interpreting and Reporting 
Evaluation Findings 

Evaluators are responsible for presenting findings to appropri- 
ate decision makers and other stakeholders and in a format 
that can be used. When different racial and ethnic groups are 


the audience of the evaluation presentation or report, the evalua- 
tor must be sensitive to their information needs (Torres, 1991). It 
is possible that various stakeholders may need some education 
about the purpose of evaluation, the methodologies that were 
used, and the potential implications of the findings. As a result, 
the expeditious and efficient use of the information by the stake- 
holders is facilitated. 

According to Torres (1991), both political and technical exper- 
tise are needed to maximize the usefulness of evaluation find- 
ings. Evaluation strategies must be integrated with organizational 
processes and political factors, and the findings must be used to 
design new programs, to alter existing ones, to reallocate resources, 
to change educational materials or strategies, and so on. Effective 
evaluation calls for the evaluator to be cognizant of factors that 
may influence a program as well as affect roles of the evaluator: 
contextual factors (e.g., the organization and its subsystems; ad- 
ministrative structure), organizational factors (e.g., size, location), 
attributes of the evaluator or program decision maker (e.g., mana- 
gerial style, biases, values), and interpersonal factors (e.g., co- 
operation, trust, conflicts). 

Knowledge of how decisions are made and the history, poli- 
tics, and economic solvency of an organization may assist in the 
appropriate use of evaluation findings (Mathison, 1991; Torres, 
1991). An analysis of these influences will aid the evaluator in 
identifying potential problem areas where the implementation of 
evaluation methodologies and the use of evaluation findings may 
occur. This analysis will also pinpoint individuals who play a 
prominent role in these influences. Knowledge and understand- 
ing of these influences will facilitate more realistic expectations 
about the nature of change within the organization. Understand- 
ing of the context, including the organizational philosophy and 
goals, also enables evaluators to determine decision-making pat- 
terns, design appropriate evaluations, and identify areas of con- 
flict and possible solutions. 

The use of periodic member checks can ensure that evalua- 
tors remain responsive to the context. This also gives members of 
the organization an opportunity to reflect on and share their own 
experiences within the organization and to develop a shared mean- 


ing of the program or organization. Understanding the perspec- 
tives of the various stakeholders is a necessary component of ef- 
fective evaluations. It can also reduce tendencies toward 
coaptation by making the evaluator aware of issues and concerns 
of the stakeholders. The beliefs and perspectives of these mul- 
tiple parties are central to the evaluation. Also important is the 
fact that evaluation outcomes have differential effects for various 
communities (Davis, 1992). For example, what works in an upper- 
middle-class suburban school may not work in a rural setting. 

Evaluation findings and their generalizations must be ethical, 
appropriate, and informed. Since culture-free research does not 
exist, it is up to the evaluator to address biases that may be present 
but are often implicit. 

The cultural biases inherent in how middle-class researchers in- 
terpret the experiences of low-income minorities may lead to 
erroneous assumptions and faulty propositions concerning causal 
relationships, to invalid social theory and consequently to in- 
valid program theory Descriptive theories derived from faulty 
premises, which have been legitimized in the literature as exist- 
ing knowledge, may have negative consequences for program 
participants. Moreover, such errors may have immediate effects 
on the lives of the minorities who are the potential beneficiaries 
of social policies. (Madison, 1992, p. 38) 

A similar sentiment is echoed by Davis (1992, p. 61) who 
stated that 

generalizations about race differences are often put forward with- 
out much attention to within-group variations and the influence 
of particular contexts. Traditionally most social prevention pro- 
grams have been conceived from dominant middle-class perspec- 
tives, and many of these programs have been implemented in 
African American communities. Some program evaluators have 
acknowledged that culturally specific approaches are needed, 
but there have been few serious efforts to design and evaluate 
programs that are based on culturally diverse perspectives. 

Multiethnocultural diversity also has implications in the re- 
porting of evaluation information. Program recipients are 


frequently members of ethnic and racial groups and tend to have 
less power and control in all phases of program implementation 
and evaluation. The decision about whose voice gets heard and 
whose gets stilled is usually made by program administrators and 
evaluators. Selection of negative versus positive comments may 
also be motivated by personal or political forces (Merithew & 
Colombo, 1992). Decisions about the distribution of evaluation 
results can also prove problematical, particularly if several differ- 
ent formats are developed. For example, the director may exer- 
cise power as the gatekeeper of information and decide that only 
the funding agency needs the full report, that the cooperating ser- 
vice agencies need the executive summary, and that program re- 
cipients do not require a report at all. These types of decisions 
should be made cooperatively with the various stakeholders; in 
this way, the differing needs of the diverse groups are more likely 
to be appropriately met. 

Evaluation requires full and complete disclosure in the report- 
ing of findings. In the case of formative evaluation, where infor- 
mation is collected to improve the delivery of services and 
activities, ongoing direct access to data is critical in order for evalu- 
ators to provide timely feedback that will help them monitor and 
fine-tune the operation of their program. Thus, their full disclo- 
sure leads to reports that are more useful to them because of the 
nature of the data that they provided to the evaluator. To allay 
any surprises of negative findings and to smooth any ethical con- 
flicts, formative evaluation data should be provided on an ongo- 
ing informal basis. These findings should also be linked with the 
specific evaluation questions that were posed (Adams, 1985). 

According to Morris and Cohn (1992), three of the four most 
frequent types of ethical problems that are encountered in an 
evaluation occur during the reporting phase: presenting the find- 
ings, misinterpreting and misusing the results, and adhering to 
the disclosure statements. A content analysis of each of these prob- 
lems revealed the following challenges to evaluators: 

Challenges in presenting the findings: (1) the evaluator felt pres- 
sured to alter his/her report because it contained information 
that the client felt uncomfortable with; (2) the evaluator is reluc- 


tant to present the findings fully for unspecified reasons; (3) an 
unethical, dangerous, or illegal behavior was encountered, and 
a decision needs to be made as to whether the evaluator should 
go public with the knowledge; and (4) the evaluator is not sure 
of his/her own ability to be neutral and balanced in presenting 
the findings. (Morris & Cohn, 1992, Table 6 on p. 12) 

Challenges in misinterpreting and misusing the results: (1) per- 
ception that the findings are "misused" or that no action was 
ever taken or that findings were used to punish, fire, or discredit 
someone, (2) plagiarism, and (3) data were altered. (Morris & 
Cohn, 1992, Table 7 on p. 12) 

Challenges in adhering to disclosure agreements: (1) there are 
disputes or uncertainties concerning ownership/distribution of 
the final report, raw data, etc.; (2) although they have not been 
pressured by the stakeholders to violate confidentiality, the evalu- 
ator is concerned that reporting certain findings could represent 
such a violation; and (3) the evaluator is pressured by the stake- 
holder to violate confidentiality. (Morns & Cohn, 1992, Table 9 
on p. 12) 

Brickell (cited in Worthen & Sanders, 1987) also identified situ- 
ations in which evaluators have experienced unethical political 
pressure: (1) the client dictates what the report will find; (2) the 
client rewrites the report; (3) the client makes future evaluation 
contracts contingent on positive findings in the current evalua- 
tion; and (4) the client introduces new requests that throw both 
the schedule and the budget off and then complains when tasks 
are not completed on time. Suggestions for dealing with these 
five political influences included the following: 

Try to understand how the client thinks. Find out what he has to 
gain or lose from the evaluation. 

Reassure the client at the outset that you can interpret the find- 
ings so as to give helpful suggestions for program improvement- 
no matter what the findings of the study are. 


Find out what the powerful decision makers — tlie client and those 
who surround him — will actually use as criteria forjudging the 
success of the project. Gather and present evidence addressed 
to those criteria. You may, if you wish, also gather data on the 
official objectives of the project or even on objectives that hap- 
pen to interest you. But never try to substitute those for data 
addressed to criteha the decision makers will use. 

Try to get a supervisory mechanism set up for the evaluation 
contract that contains a cross section of all the powerful deci- 
sion makers. Try to get it designed so that the members have to 
resolve the conflicts among themselves before giving you march- 
ing orders for the study or deciding whether to accept your final 

Write the report carefully especially when describing shortcom- 
ings and placing blame, and do mention any extenuating cir- 
cumstances. Review the draft final report before submitting it to 
the client for his review, making sure in advance that you can 
defend any claim you make. (Brickell, cited in Worthen & Sand- 
ers, 1987, p. 295) 

The report itself may also cause problems for participants. 
Evaluators need to be aware of the potential effects of an evalua- 
tion on the unempowered groups that the programs are intended 
to aid. Davis (1992, p. 64) suggested the following to ensure ap- 
propriate evaluation results and interpretations when race is used 
as a variable: 

• Look at within-group variations as well as between- 
group differences; 

• Use evaluation teams consisting of members of ethnic/ 
racial communities; 

• Have members of ethnic /racial groups review the evalua- 
tion results prior to dissemination; 

• Collect information about the social location of the pro- 
gram since the context can inform the results and inter- 
pretations of an evaluation study; and 

• Examine different conceptual meanings of race to see 
whether evaluation interpretations are affected. 


Efforts must be taken to ensure that the report is being used 
as it was originally planned (Covert et al, 1988). Its acceptability 
will depend on whether it presented findings in a format conducive 
to its use, whether the information was credible, whether the report 
answered the questions that were originally posed, and whether it 
was received on time (Chelimsky 1987). The dissemination of the 
evaluation report must adhere to the protocols that were set forth 
in the evaluation contract. It is also recommended that this distri- 
bution be confirmed with relevant stakeholders in advance. 

Adams (1985) suggested that a credible external evaluator be 
hired to critique the evaluation report as well as the design. This 
self-disclosure can mitigate any problems that may arise about 
the report itself due to the unfamiliarity of stakeholders about the 
purpose and process of evaluation. In addition, the evaluator 
should consider having diverse audiences review the evaluation 
report for cultural sensitivity. 

Role of the Evaluator 

Definition of the role of the evaluator depends on the organiza- 
tional context and the position that the evaluator holds within 
the organization. Some researchers have made the distinction be- 
tween the internal and external evaluator, where the internal evalu- 
ator is a member of the staff and the external evaluator is a member 
of an outside organization. An area of potential conflict between 
the internal evaluator and the client is that the organization ex- 
pects the evaluator to work in its best interest. This may mean 
that the internal evaluator has to weigh organizational loyalty and 
professional objectivity (Adams, 1985). Mathison (1991) observed 
that coaptation is a particularly salient problem for internal evalu- 
ators because of the role conflicts associated with being a profes- 
sional evaluator, a member of a substantive field, and a member 
of the organization being evaluated. Because of the nature of or- 
ganizations to maintain the status quo, sensitivity to ethical di- 
lemmas is thus diminished. For the internal evaluator, the issue 
of maintaining independence from decision makers looms large. 
Adams (1985) suggested that internal evaluators establish 
an internal support base. This can be accomplished through a 


variety of means, including meeting with key stakeholders peri- 
odically and attending both formal and informal activities. These 
strategies increase the visibility of the evaluator and also provide 
an opportunity for stakeholders to communicate issues of con- 
cern to the evaluator. 

Complacency about program functions and operations can be 
minimized through sensitive questioning and open discussions 
with stakeholders. Torres (1991, pp. 190-191) advocated that 

management should take into account individual values, per- 
spectives, and needs for growth.... As a consultant-mediator, 
the internal evaluator collects, interprets, and reports informa- 
tion which provides leaders opportunities to reflect critically on 
the organizational context and the individuals who comprise it. . . . 
A fundamental task for internal evaluators is to avoid coaptation 
and habitual responsiveness to management concerns, while ad- 
dressing the larger goal of representativeness in decision mak- 
ing. Doing so requires: (1) understanding contextual influences 
on the practice and use of evaluation within an organization, as 
well as on all other major organizational goals and operational 
area; (2) identifying and understanding the perspectives of stake- 
holders; (3) working to maximize credibility and trust with all 
constituents; (4) aligning methods with the epistemological ori- 
entations of the evaluation audiences; (5) raising and represent- 
ing issues to those in authority; (6) educating management on 
the relationship between their perspectives and the perspectives 
of others; and (7) maintaining a tolerance for ambiguity and 
incremental change. 

Morris and Cohn (1992) found that ethical dilemmas were ex- 
perienced by external evaluators more often than internal evalu- 
ators. They speculated that this may be due to internal evaluators 
too closely identifying with the client and the program. On the 
flip side, while external evaluators may find it easier to maintain 
objectivity, they may also be too removed from the program to 
understand all the nuances of program delivery, the interpersonal 
dynamics, and political intrigues. It is more difficult to hide nega- 
tive consequences of programs or incompetent staff members from 
internal evaluators, since they are aware of program operations 


on a more intimate basis. However, these negative realities do not 
necessarily pose ethical problems for them in their role as evalua- 
tors (Morris & Cohn, 1992). 

At a recent national conference, Greene (1992) wondered 
whether the evaluator could ethically intervene in a program with 
personal observations and suggestions, while Bernstein (1992) 
asked what would happen if the evaluator disagreed with the 
intervention selected. One could take this a step further to query 
what the proper role is for evaluators where moral issues are con- 
cerned (Smith, 1985a). Schwandt (1992) suggested an alternative, 
whereby the role of the evaluator was "neither adopt[ing] the 
dominant value perspectives of the client, nor promot[ing] some 
higher moral ground, acting as a philosopher king. The notion of 
scientific (or moral) expert gives way to social commentator or 
critic exploring the moral and political meaning of programs with 
stakeholders. The evaluator seeks to foster the posture of a reflec- 
tive practitioner among stakeholders and is far more concerned 
with probing rather than proving'' (Schwandt, 1992, p. 141). 

Hendricks (1985) also proposed that one of the roles of evalu- 
ators is to question the appropriateness of services when con- 
fronted with questionable effects or implications of certain policies 
or procedures or both. To him, an evaluation is incomplete if it 
does not examine whether the services are congruent with values 
embodied in their objectives. However, certain risks accompany 
the decision to judge appropriateness — the decision maker might 
view these judgments to be his or her domain; the decision maker 
may feel threatened and "punish'' the evaluator (e.g., firing, re- 
fusing to cooperate in future evaluation activities, sabotaging the 
evaluator 's work with other stakeholders); the evaluator may 
appear to be morally superior to the decision maker; and the evalu- 
ator may lose credibility for being objective. 

The credibility of the researcher is a decisive factor in using 
evaluation results, particularly in qualitative inquiry where the 
primary data collection instrument is the investigator. To address 
this issue, Patton (1990) suggested that "the principle is to report 
any personal and professional information that may have affected 
data collection, analysis, and interpretation — either negatively or 
positively — in the minds of the users of the evaluation" (p. 472). 


Sieber and Sanders (1978, p. 118) advocated that "an actively in- 
volved evaluator can, and should, press for consideration of other 
information than that originally contracted for if it becomes ap- 
parent that such information provides insight into the decision 
context at hand.... While the purpose of evaluation is always to 
provide information for decision makers, the character of the 
evaluator 's role(s) should depend on the decision context and the 
decision maker's information needs.'' 

Training of the Evaluator 

The adequate and appropriate training of evaluators has great 
implications for the field. Smith (1985b) has suggested that evalu- 
ators receive training on professional and program moral issues 
and problems, including a discussion of existing ethical profes- 
sional codes and case studies that illustrate the nature, frequency, 
and severity of ethical problems. A review of studies (e.g., Morris 
& Cohn, 1992; Newman & Brown, 1992; Pope & Vetter, 1992) about 
ethical dilemmas faced by evaluators could serve as a useful start- 
ing point. Special attention should be paid to developing and 
improving the communication and interpersonal skills of evalua- 
tors since the collection of complete evaluation data relies on the 
rapport that evaluators develop with their clients. "The three com- 
munication skills that appear to have the greatest potential for 
smoothing ethical conflicts are (1) involving key decision makers 
in designing the evaluation; (2) sharing findings incrementally 
and informally so that negative findings are not so surprising; 
and (3) relating findings to specific issues of interest to decision 
makers" (Adams, 1985, p. 56). Because of the frequently sensitive 
nature of data, emphasis on ensuring confidentiality of data and 
discreetness on the part of the evaluator is needed. According to 
Torres (1991), the credibility of an evaluation report depends not 
only on the evaluator 's competence and technical skills, but also 
on his or her ability to maintain confidentiality and to work coop- 
eratively with stakeholders. 

Lycke (1992) advocated the use of mentors in developing ethi- 
cal competence in professionals since they play an important role 
as models or socializing agents. Mentors also promote an increase 


in professional reflection on ethical issues where the relationship 
between professional practice and practical theory is discussed and 
debated. She outlined the following as possible roles of a mentor 
(Lycke, 1992, p. 4): 

• To model ethical conduct and judgment by serving as a 
model to others in his or her dedication, interest, and 
thoroughness; having an integrated knowledge of the 
professional code of ethics and local standards; making 
decisions that are "right'' in the collective opinions of 

• To know how to reason and justify ethical judgments by 
identifying the important facts, the salient values at stake 
and those in conflict, and the relevant and defendable 

• To be sensitive to his or her own and others' reasons, 
justifications, and judgments by understanding the 
interests of others and their arguments 

• To be able to assist others in learning how to make ethical 
judgments and to substantiate these with arguments and 
also by perceiving the frustrations of the protege dealing 
with conflicting or unstated codes 

Additionally, Lycke (1992) proposed that three levels of prac- 
tice be addressed: the actions considered, the experience and 
knowledge on which they are based, and the values and ethical 
justifications involved. E valuators need to consider the norms that 
affect ethical issues, such as laws, rules, regulations, declarations, 
conventions, professional codes of ethics, and personal ethics. 

In addition, professional evaluation organizations can serve 
as a forum for discussing ethical issues. A discussion of fieldwork 
dilemmas not only provides the opportunity to share problems, 
debate potential solutions, and propose future directions, but also 
furnishes the impetus for change within the profession. Profes- 
sional codes of ethics should be scrutinized and debated since 
they define the potential conduct of the professional, serve as a 
means of self-regulation within the profession, or represent the 
profession's desire to minimize external regulation. For example, 
the standards issued by the Joint Committee on Standards for 


Educational Evaluation are subjected to a critical review accord- 
ing to a specific set of protocols every 10 years. After a national 
review, the 1981 standards were reorganized and reissued in 1994. 
These discussions can be illuminating since professional codes of 
ethics cannot regulate how researchers should act in all possible 
circumstances (Diener and Crandall, 1978; Lycke, 1992; Taylor, 1987). 

Training of evaluators must also address multiethnocultural 
issues and concerns. Orlandi (1992b) proposed a cultural sophisti- 
cation framework, with a continuum going from culturally incompe- 
tent (with an overall effect of being ''destructive'') to culturally 
sensitive (with an overall effect of being "neutral") to culturally 
competent (with an overall effect of being "constructive"). The goal 
of training programs would be to enhance cognitive development, 
affective development, and skills development as they relate to 
cultural competence. 

One framework for examining the moral aspects of a program 
and its evaluation is to look at the dominant values evident in the 
program at each stage of its development, implementation, and 
evaluation. Smith (1985b, p. 8) identified a series of questions to 
ask at each stage: 

• Program as intended. What values are designed into the 
program or on what values is the program based? 

• Program as operated. What do the people operating the 
program value? What values are being supported or 
fostered by the program? 

• Program as received. What do the people receiving or 
experiencing the program value? Which of their values 
are supported or undermined by the program? 

• Program as evaluated. What do the evaluators value or on 
what values is the evaluation based? What do these 
people wanting the program evaluated value? What 
values are used to judge the program? 

Developmental Evaluation Model 

Of particular value in the ethical design of evaluations for the 
multiethnocultural community is a model developed by Stockdill 


et al. (1992). Known as the developmental evaluation model 
(DEM), its goal is to bring representatives of the varying groups 
and stakeholders into the process of evaluation from its early stage 
of goal setting and program planning. Special emphasis was 
placed on involving people of various cultures and racial /ethnic 
backgrounds and adding their perspective to the stakeholder in- 
volvement processes. The approach calls for a careful examina- 
tion of the values and norms of each targeted group and builds 
these unique cultural perspectives into the design of a commu- 
nity-based intervention and its evaluation. The DEM was devel- 
oped when Stockdill and her colleagues were engaged in the 
evaluation of a multicultural education program. Supporting Di- 
versity in Schools (SDS), where there were five racial and ethnic 
groups: Native Americans, African Americans, Hmong Americans, 
Cambodian Americans, and Hispanic /Latino Americans. These 
evaluators wanted ''to create a shared evaluation process that is 
sensitive to and empowering of people of color and respectful of 
their diverse perspectives'' (Stockdill et al., 1992, p. 17). 

For the process to work, for it to be genuinely empowering and 
bottom-up, tfie local community must struggle with setting its 
own ends and working to accomplish them.... The evaluator 
can become an enabling partner in the process, supporting the 
local community program staff members, and school people in 
discovering what they want to accomplish and learning how to 
articulate claims about the differences they are committed to 
making. Some of these elements are similar to those found in 
formative evaluation, but there are important differences. For- 
mative evaluation typically assumes that ultimate goals are 
known, and that the issue is how best to reach them. By con- 
trast, developmental evaluation is preformative in the sense that 
it is part of the process of developing goals and implementation 
strategies. Developmental evaluation brings the logic and tools 
of evaluation into the early stages of community, organization, 
and program development. (Patton, 1992, p. 26) 

One of the unique features of their model was to get the par- 
ticipants to make claims statements regarding SDS at its outset. 
These claims represented changes that the participants were 


committed to make as a result of activities conducted by the SDS. 
Based on principles of consensus in decision making and the 
meaningful involvement of the various stakeholders, the effec- 
tive use of claims is critical for the successful implementation of 
the program. These claims must be developed and owned by those 
responsible for the program's activities and impacts throughout 
the community and the schools rather than by a small group of 
administrators and evaluators. 

In a developmental approach, the articulation of goals, claims, 
and supporting measures is one of the outcomes of the process, 
rather than one of the foreordained determinants of the pro- 
cess. Moreover, the goals, claims, and measures that develop 
during the process may change as the explorers of the new ter- 
ritory come to a better understanding of the lay of the land. 
When clear, specific, measurable goals are established at the 
moment a grant is made, the struggle of community people to 
determine their own ends is summarily preempted, and they are 
once again disempowered — this time in the name of evaluation. 

The logic of evaluation can be a powerful force for helping people 
clarify their thinking and become rigorous in holding themselves 
accountable. The highest form of accountability is self-account- 
ability. If the community learns to hold itself accountable for its 
own aims, that is much more empowering than to be held ac- 
countable by external funders. This means that much of the evalu- 
ation activity in a developmental approach involves training local 
people to use evaluation logic and helping them develop their 
own goals, claims, criteria, and methods. 

This kind of process clearly takes time. Much evaluation meth- 
odology assumes the accomplishment of concrete, measurable 
outcomes in one or two years, but community development pro- 
cesses take five to ten years. In our rapidly changing world, it 
would be absurd to force immediate commitment to the un- 
known goals and criteria of a five- to ten-year process. It is pos- 
sible to make some commitments and specify some things (for 
example, active involvement of diverse groups of people, or on- 
going articulation of developmental goals by participants in the 


process), but even these must be negotiated as part of the pro- 
cess. (Patton, 1992, pp. 26-27) 

In addition, the DEM relies on local residents to assist in the 
evaluation. These multicultural evaluation liaisons receive train- 
ing that will facilitate the collection of data on program activities, 
outcomes, and claims. Thus, the DEM is one possible strategy for 
working in a multiethnocultural community. 


In summary, this chapter has focused on the importance of be- 
coming cognizant of ethical issues that evaluators must address, 
particularly in multiethnocultural settings. A distinction was made 
between evaluation moral issues and program moral issues, and 
the need for evaluators to address both. Ethical conflicts that are 
frequently encountered in each stage of the evaluation process — 
from the development of the contract to the reporting of the find- 
ings — ^were also identified and discussed. The implications of these 
ethical dilemmas were presented in the context of multiethno- 
cultural settings. As evaluators, we would do well to follow 
Madison's rationale for why evaluators should attend to program 
moral issues where people of various cultures and racial /ethnic 
backgrounds are the recipients of social programs: 

As a social practice, evaluation entails a public responsibility. This 
public responsibility should include ascertaining the truth about 
programs. Evaluators have a responsibility to those most affected 
by the program under review to use every method available for 
seeking the truth. The ultimate goal of evaluators should be to 
ascertain the closest approximation of truth about the impact of 
social policy on the real world of program participants. (Madi- 
son, 1992, p. 41) 


The author gratefully acknowledges Michael Morris and James 
E. Tokley, Sr., for their review of this manuscript. Their insightful 
comments led to the final revision and are greatly appreciated. Ronald 
L. Braithwaite was also key in his encouragement and support. 



Adams, K.A. (1985). Gamesmanship for internal evaluators: Knowing when to 
"hold 'em" and when to "fold 'em." Evaluation and Program Planning, 8, 53-57. 

Beauchamp, T.L., & Childress, J.F. (1983). Principles of biomedical ethics. New York: 
Oxford University Press. 

Bernstein, D.J., (1993, November). Some thoughts about ethics and evaluation: A 
practitioner's theoretical perspective. Paper presented at the annual meeting of 
the American Evaluation Association, Seattle, WA. 

Bunda, M.A. (1985). Alternative systems of ethics and their application to 
education and evaluation. Evaluation and Program Planning, 8, 25-36. 

Cassell, J. (1982). Harms, benefits, wrongs, and rights in fieldwork. In J.E. Sieber 
(Ed.), The ethics of social research: Fieldwork, regulation, and publication (pp. 7-31). 
New York: Springer- Verlag. 

Chelimsky, E. (1987). The politics of program evaluation. In D.S. Cordray, H.S. 
Bloom, & R.J. Light (Eds.), Evaluation practice in review: Vol. 34. New directions 
in program evaluation (pp. 5-21). San Francisco: Jossey-Bass. 

Covert, R., Voorhees, M., & Honea, G.E. (1988, October). Planning for ethical inquiry. 
Workshop presented at the annual meeting of the American Evaluation 
Association, New Orleans, LA. 

Davis, J.E. (1992). Reconsidering the use of race as an explanatory variable in 
program evaluation. In A.M. Madison (Ed.), Minority issues in program 
evaluation: Vol. 53. New directions for program evaluation (pp. 17-33). San 
Francisco: Jossey-Bass. 

Diener, E., & Crandall, R. (1978). Ethics in social and behavioral research. Chicago: 
University of Chicago Press. 

Ericson, D.P. (1990). Social justice, evaluation, and the educational system. In 
K.A. Sirotnik (Ed.), Evaluation and social justice: Issues in public education: 
Vol. 45. New directions for program evaluation (pp. 5-21). San Francisco: Jossey- 

Fang, W.L. (1986, October-November), from learning to practice: Taking photographs 
of the evaluation conference. Paper presented at the annual meeting of the 
American Evaluation Association, Kansas City, MO. 

Fang, W.L., & Ellwein, M.C. (1990). Photography and ethics in evaluation. 

Evaluation Review, 14, 100-107. 

Gold, S.J. (1989). Ethical issues in visual field work. In G. Blank, J.L. McCartney, 
& E. Brent (Eds.), New technology in sociology: Practical applications in research 
and work. New Brunswick: Transaction Books. 

Grace, C.A. (1992). Practical considerations for program professionals and 
evaluators working with African- American communities. In Mario A. Orlandi 


(Ed.), Cultural competence for evaluators: A guide for alcohol and other drug abuse 
prevention practitioners working with ethnic/racial communities (pp. 55-74). 
Rockville, MD: U.S. Department of Health and Human Services. 

Greene, L. (1992, November). Invited speaker for the Topical Interest Group on Health 
Evaluation. Paper presented at the annual meeting of the American Evaluation 
Association, Seattle, WA. 

Hendricks, M. (1985). Should evaluators judge whether services are appropriate? 
Evaluation and program planning, 8, 37-4A. 

House, E.R. (1990). Methodology and justice. In K.A. Sirotnik (Ed.), Evaluation 
and social justice: Issues in public education: Vol. 45. New directions for program 
evaluation (pp. 5-21). San Francisco: Jossey-Bass. 

Jerrell, S.L., & Jerrell, J.M. (1985). Road signs in ethical quicksand. Evaluation and 
Program Planning, 8, 73-76. 

Joint Committee on Standards for Educational Evaluation. (1994). The program 
evaluation standards: How to assess evaluations of educational programs. Thousand 
Oaks, CA: Sage Publications. 

Kimmel, A.J. (1988). Ethics and values in applied social research. Applied Social 
Research Methods Series, 12, 1-160. 

Lincoln, Y.S., & Cuba, E.G. (1986, October-November). Ethics: The failure of 
positivist science. Paper presented at the annual meeting of the American 
Evaluation Association, Kansas City, MO. 

Linney, J.A., & Wandersman, A. (1991). Prevention Plus III: Assessing alcohol and 
other drug prevention programs at the school and community level. Rockville, MD: 
Office for Substance Abuse Prevention, U.S. Department of Health and Human 

Lycke, K.H. (1992). The mentor role in promoting ethical considerations in 
professional practice. Professions Education Research Quarterly, 14(1), 2-5. 

Madison, A.M. (1992). Primary inclusion of culturally diverse minority program 
participants in the evaluation process. In A.M. Madison (Ed.), Minority issues 
in program evaluation: Vol. 53. New directions for program evaluation (pp. 35-43). 
San Francisco: Jossey-Bass. 

Mathison, S. (1991). Role conflicts for internal evaluators. Evaluation and Program 
Planning, 14, 173-179. 

Merithew, M., & Colombo, M. (1992, November). Searching for ethical procedures 
to guide the use of direct quotations in evaluation reports. Paper presented at the 
annual meeting of the American Evaluation Association, Seattle, WA. 

Morris, M., & Cohn, R. (1992, November). Program evaluators and ethical challenges: 
A national survey. Paper presented at the annual meeting of the American 
Evaluation Association, Seattle, WA. 

Newman, D.L., & Brown, R.D. (1992). Violations of evaluation standards: 
Frequency and seriousness of occurrence. Evaluation Review, 16, 219-234. 


Orlandi, M.A. (1992a). The challenge of evaluating conununity-based prevention 
programs: A cross-cultural perspective. In M.A. Orlandi (Ed.), Cultural 
competence for evaluators: A guide for alcohol and other drug abuse prevention 
practitioners working with ethnic/racial communities (pp. 1-22). Rockville, MD: 
U.S. Department of Health and Human Services. 

Orlandi, M.A. (1992b). Defining cultural competence: An organizing framework. 
In M.A. Orlandi (Ed.), Cultural competence for evaluators: A guide for alcohol and 
other drug abuse prevention practitioners working with ethnic/racial communities 
(pp. 293-299). Rockville, MD: U.S. Departinent of Health and Human Services. 

Patton, M.Q. (1985). Cross-cultural nongeneralizations. In M.Q. Patton (Ed.), 
Culture and evaluation: Vol. 25. New directions for program evaluation (93-96). 

Patton, M.Q. (1990). Qualitative evaluation and research methods. Newbury Park, 
CA: Sage Publications. 

Patton, M.Q. (1992). Developmental evaluation: An approach for empowerment- 
oriented and multicultural programs. In A.M. Madison (Ed.), Minority issues 
in program evaluation: Vol. 53. New directions for program evaluation (pp. 17-33). 
San Francisco: Jossey-Bass. 

Payne, D. A. (1994). Designing educational project and program evaluations: A practical 
overview based on research and experience. Boston: Kluwer Academic Publishers. 

Pope, K.S., & Vetter, V.A. (1992). Ethical dilemmas encountered by members of 
the American Psychological Association: A national survey. American 
Psychologist, 47, 397-411. 

Ramos, M.C. (1989). Some ethical implications of qualitative research. Research 
in Nursing and Health, 12, 57-63. 

Schwandt, T.A. (1992). Better living through evaluation? Images of progress 
shaping evaluation practice. Evaluation Review, 13, 135-144. 

Shadish, W.R., Jr., Cook, T.D., & Leviton, L.C. (Eds.) (1991). Foundations of program 
evaluation: Theories of practice. Newbury Park, CA: Sage Publications, Inc. 

Shadish, W.R., Jr., & Epstein, R. (1987). Patterns of program evaluation practice 
among members of the Evaluation Research Society and Evaluation Network. 
Evaluation Review, 11, 555-590. 

Sheinfeld, S.N., & Lord, G.L. (1981). The ethics of evaluation researchers: An 
exploration of value choices. Evaluation Review, 5, 377-391. 

Sieber, J.E. (1980). Being ethical: Professional and personal decisions in program 
evaluation. In R. Perloff & E. Perloff (Eds.), Values, ethics, and standards in 
evaluation: Vol. 7. New directions for program evaluation. San Francisco: 

Sieber, J.E. (1992). Planning ethically responsible research: A guide for students 
and internal review boards. Applied social research methods series: Vol. 31 (pp. 
1-163). Beverly Hills, CA: Sage Pubhcations. 


Sieber, J.E., & Sanders, N. (1978). Ethical problems in program evaluation: Roles, 
not models. Evaluation and Program Planning, 1, 117-120. 

Smith, N.L. (1985a). Introduction: Moral and ethical problems in evaluation. 

Evaluation and Program Planning, 8, 1-3. 

Smith, N.L. (1985b). Some characteristics of moral problems in evaluation 
practice. Evaluation and Program Planning, 8, 5-11. 

Spradley, J.P. (1979). The ethnographic interview. Fort Worth, TX: Holt, Rinehart 
and Winston, Inc. 

Stake, R.E. (1986). Quieting reform: Social science and social action in an urban youth 
program. Urbana, IL: University of Illinois. 

Stake, R.E. (1991). Responsive evaluation and qualitative methods. In W.R. 
Shadish, T.D. Cook, & L.C. Leviton (Eds). Foundations of program evaluation: 
Theories of practice. Newbury Park, CA: Sage Publications. 

Stockdill, S.H. (1992). Implementing the developmental evaluation approach. 
In A.M. Madison (Ed.). Minority issues in program evaluation: Vol. 53. New 
directions for program evaluation (pp. 17-33). San Francisco: Jossey-Bass. 

Stockdill, S.H., Duhon-Sells, R.M., Olson, R.A., & Patton, M.Q. (1992). Voices in 
the design and evaluation of a multicultural education program: A 
developmental approach. In A.M. Madison (Ed.) Minority issues in program 
evaluation: Vol. 53. New directions for program evaluation (pp. 17-33). San 
Francisco: Jossey-Bass. 

Taylor, S.J. (1987). Observing abuse: Professional ethics and personal morality 
in field research. Qualitative Sociology, 10, 288-302. 

Thome, B. (1980). You still takin' notes? Fieldwork and problems of informed 
consent. Social Problems, 27, 284-297. 

Torres, R. (1991). Improving the quality of internal evaluation: The evaluator as 
consultant-mediator. Evaluation and Program Planning, 14, 189-198. 

Worthen, B.H., & Sanders, J.R. (1987). Educational evaluation: Alternative approaches 
and practical guidelines. New York: Longman Inc. 



Evaluating the Success of 

Community-Based Substance 

Abuse Prevention Efforts: 

Blending Old and New 

Approaches and Methods 

Sehwan Kim, Ph.D., Barry Kibel, Ph.D., 
Charles Williams, Ph.D., and Nancy Hepler, Ph.D. 


Nothing short of an institutional revolution is afoot in the sub- 
stance abuse prevention arenas of the Nation. Community and 
organizational coalition activity has become the principal force 
for coordinating and mobilizing efforts aimed at curtailing sub- 
stance abuse. A recent study conducted by Pacific Institute for 
Research and Evaluation for Join Together (a national resource 
for communities fighting substance abuse funded by The Robert 
Wood Johnson Foundation) found coalition-based prevention ac- 
tivities under way in every State and Territory, all 25 of the Nation's 
largest cities, and literally thousands of communities of every size 
(Join Together, 1992). Sparking the revolution are private founda- 
tions such as Robert Wood Johnson and Federal agencies such as 
the Center for Substance Abuse Prevention (CSAP), the latter pro- 
viding 5-year grants for partnership activities to more than 250 
local coalitions. 


As reported in the Join Together survey, most coalitions form 
in response to the realization that scattered, uncoordinated ac- 
tions of public and private agencies are not enough. Faced with a 
real need to work together, as well as financial incentives to do so, 
coalitions have been created that comprise a far broader range of 
organizations, institutions, and community groups than have pre- 
viously been involved in substance abuse reduction activities. The 
majority of coalitions boast representation and shared leadership 
of professionals, large organizations, lay people, activists, and 
government officials. Schools, local law enforcement, prevention 
and treatment providers, courts /probation offices, parents and 
religious organizations, health service agencies, and volunteers 
are present on a majority of the coalitions surveyed. Volunteers, 
in particular, provide a critical resource for sustaining the activi- 
ties of these partnerships. 

For a growing number of these coalitions, however, the "hon- 
eymoon period" is nearly over. Getting people from different pro- 
fessional, racial, cultural, religious, economic, and institutional 
backgrounds to sit around the table and work together has been — 
and continues to be — a formidable challenge. However, the true 
value of these coalitions must necessarily be gauged by the suc- 
cess of programs, services, and policy changes they catalyze that 
collectively have the desired impact on substance abuse problems 
in their respective communities. It is no simple matter to produce 
such impacts. 

There are very few single interventions (e.g., program, ser- 
vice, policy) that alone can produce a significant substance abuse 
impact. A dramatic example of an apparent exception was the 
Federal pressure put on States in the mid-1980s to raise the legal 
drinking age to 21 in order to avoid losing highway funds. Sig- 
nificantly fewer crashes and deaths involving youth now occur 
on U.S. highways because of this intervention. But even here the 
intervention's effects were buttressed by complementary inter- 
ventions such as the increased existence and enforcement of seat 
belt laws. Efforts of Mothers Against Drunk Driving, Students 
Against Drunk Driving, and other groups to discourage drinking 
and driving may have contributed to the impact, general down- 


ward shifts in alcohol consumption, coupled with school-based 
education programs, may have contributed further. 

Coalitions will not find "magic bullets'' with which to arm 
their communities in the fight against substance abuse. Commu- 
nities are being challenged to improvise strategies involving mul- 
tiple, complementary, and result-reinforcing interventions. To 
gauge their success, as well as to justify their continued existence, 
the coalitions need to know what interventions are working, sin- 
gly and in combination with one another. 

Local and national evaluators are being called upon to help pro- 
vide this essential information. To be responsive to the call, evalua- 
tors are recognizing the need to become partners within the coalitions. 
In an arena of give-and-take, learning by doing, and ''radical incre- 
mentalism" (a term evoked by Tom Peters [1987] to describe a pro- 
cess of constant experiments aimed at implementing a revolutionary 
future), there may be no appropriate place for a sideline observer. 

In this chapter, we examine the new tools-of-the-trade of the 
"collaborative evaluator." Included among these are some famil- 
iar implements as well as some newer techniques adapted from 
Total Quality Management. The chapter has two major parts. Part 
I explores the basic tenets and approaches of "outcome-focused 
evaluation," where new roles for the collaborative evaluator are 
being called into play. Part II offers a bridge back to the tradi- 
tional research tools that remain invaluable for checking assump- 
tions, testing hypotheses, gauging outcomes, and determining 
which factors contribute to success in various contexts. The chap- 
ter concludes with a brief return to the challenge facing coalitions 
in meaningfully engaging their network of partners. Again, a 
unique role for evaluators in this process is highlighted. 

Part I. Outcome-Focused Evaluation 

The Need for Expanding the Role 
of Evaluators 

Evaluators today are being presented with a unique challenge. 
Not only are most coalitions required by their funding sources to 


integrate the evaluation function into their programs, but the coa- 
litions are also in need of the types of feedback that effective evalu- 
ation can provide. To be useful, however, evaluators must supplement 
their traditional approaches and methodologies with new think- 
ing and new forms of action. Note that we are not suggesting a 
wholesale abandonment of the approaches and tools that have 
long supported the field. Rather, we are suggesting that these 
approaches and tools are simply not enough. There are four rea- 
sons for this assertion. 

First, as already discussed, there is the widely accepted rec- 
ognition that substance abuse problems demand coordinated and 
multiple, rather than isolated and singular, intervention strate- 
gies. With multiple, simultaneously executed interventions, it may 
not be possible (even with complex systems-simulation models) 
to isolate the unique contributions of any single intervention. 

Second, although several communities may elect to implement 
the same mix of interventions, they will certainly execute them in 
different ways. Hence, there is little hope of compiling a large 
sample of comparable multiple-intervention experiments to be 
assessed against a sample of alternative approaches. 

Third, the communities being studied are dynamic, open sys- 
tems that are continually being exposed to "externar' influences. 
Often such influences will confound and even dwarf the effects 
that might otherwise be attributed to the local interventions be- 
ing studied. The evaluator must therefore be vigilant toward these 
influences and, where possible, incorporate their effects within 
the evaluation. 

Fourth, along with coalition building comes the renewed as- 
sertion of multicultural values and appreciation for open, cultur- 
ally sensitive dialog. The detached observer, silently drawing 
private conclusions about the activities of the coalition, is likely 
to be viewed as anathema to the process and approached with 
distrust and disdain. 

Reviewing this list, it is hard to see where the classical experi- 
mental model (based on a single, isolated intervention applied in 
a controlled environment with well-matched controls) fits. In fact, 
there is a definite role for this model within the totality of the 
evaluation effort. This role will be explored in Part 11. Here, how- 


ever, we will focus on the role that evaluators can perform in dy- 
namic, "noisy,'' collaborative environments, where several comple- 
mentary experiments are under way simultaneously Such 
situations are unlikely to be replicated elsewhere in precisely the 
same ways. 

Back to the Basics 

To begin, let us briefly review some of the basic tenets of ''good'' 
evaluation as espoused by two of the acknowledged leaders in 
the field, Ernest House (1992) and Peter Rossi (1992), in their ad- 
dresses to plenary sessions of the conference of the American 
Evaluation Association. Ernest House stressed that methodology 
is productive only if the content of the evaluation is considered 
first and then content and methodology are creatively married. 
Too frequently, he noted, methodology is in\posed on content to 
the detriment of both. 

Evaluation, according to House, is an assessment of the merit 
or worth of something (e.g., a program, job performance, public 
policy) when measured against acceptable criteria of excellence. 
A task for the evaluator is to work with the client to establish the 
criteria that will serve as the bases, as well as set the tone, for the 
evaluation.^ The complexity of the criteria may demand the use 
of multiple methodologies for observation and assessment. The 
findings obtained may reflect a mix of rigorous inquiry, common 
sense, and experience-based reasoning — all of which may be ap- 
propriate. The key in combining such methods, according to 
House, is establishing some model or system of evaluation coor- 
dinates upon which to collectively map the findings being gath- 
ered. This allows the telling of a comprehensive, useful story. 

^This is a very important point. Consider a situation in your own past 
when a teacher or supervisor gave you an unfair performance evalu- 
ation. More likely than not, you would not have complained about 
the 'Tacts" of the evaluation, but about the criteria used. You might, 
for example, have pointed out that you made a major contribution 
to the organization, or an accomplishment in the class, that was com- 
pletely missed — that too much of the evaluation focused on sec- 
ondary issues that were ultimately of no significance. 


Peter Rossi (1992) reinforced House's assertions. He argued 
that programs being evaluated cannot be treated as "black boxes'' 
(i.e., one cannot simply look at interventions and outcomes with- 
out studying the complex transformation processes linking them). 
The evaluator has to become intimately involved with a program 
to grasp why it is working or not working — and how to fix it 
when it does not work. 

Rossi noted that the context and scale of the program most 
often dictate the types of methodology that are most useful. Large 
national programs, for example, demand a radically different ori- 
entation on the part of the evaluator than small community- 
focused programs. Likewise, multisite evaluations call for 
strategies different from those appropriate for single-site studies. 
The interests and requirements of those who fund the evaluation 
further focus the nature of that evaluation. Locally funded eval- 
uations are generally expected to provide more immediate, 
prograra-focused feedback than are nationally funded studies, the 
latter tending to involve more quantitative analyses in search of 
valid cost-benefit measures. Middle-sized evaluations, Rossi ob- 
served, often lead to conceptual struggles between the emphasis 
on quantitative or qualitative evaluation. He suggested that the 
needs of the funding sources should dictate whether a more quan- 
titative or more qualitative response is called for. 

Let us now turn back to coalition-based prevention efforts to 
see how these basic principles might be applied. 

The Multiple-Gates Model 

The success of a community-based coalition engaged in substance 
abuse prevention will reflect the contributions of a growing set of 
interventions catalyzed by the coalition and the collective impact 
of these interventions on reducing such abuse. A critical challenge 
for the coalition is to maximize the unique contributions of each 
of these interventions toward abuse prevention. In this context, 
the evaluator is entrusted with serving the role of "honest score- 
keeper." The scorekeeper maintains a running tally of progress of 
coalition-catalyzed efforts, since the coalitions need to know what 
is working and why (and what is not working and why). The 


evaluator is further called upon to serve as the ''play-by-play ana- 
lyst/' The analyst pinpoints areas of strength and weakness in 
project design, in project execution, or in the operations of the 
coalition. Finally, to translate insights regarding what works and 
why into practical advice for the coalition, the evaluator also needs 
to assume the role of "assistant coach/' The coach offers sugges- 
tions for mid-course corrections, additional interventions, and 
possible shifts in the strategic thrust of the coalition. 

These three roles (i.e., evaluation functions) may seem to be 
in conflict. How can the evaluator retain the "objectivity" required 
of a scorekeeper while also serving as assistant coach? How can 
the evaluator remain value-neutral in reporting and assessing 
what is transpiring when the activities may reflect advice the 
evaluator has offered to the coalition? The key to resolving these 
apparent conflicts lies in basing the evaluation on established 
performance criteria and progress indicators. 

To explain, consider the model developed by Allan Cohen and 
Barry Kibel (Pacific Institute for Research and Evaluation, 1992) 
to delineate the successive stages in implementing an interven- 
tion (see Figure 1). The justification for a particular intervention 
emerges from a community or organization's mission (box 1) and 
the associated strategic (box 3) and tactical (box 5) planning pro- 
cesses used to translate that mission to future actions. Once the 
decision is reached regarding what action is called for, the inter- 
vention is designed in detail (box 7). Then it is executed. Cohen 
and Kibel then differentiate three hierarchically distinct stages of 
effects stemming from the intervention: 

• Immediate effects (box 9). The direct, short-term influence 
of the intervention on those for whom it is designed to 
serve or benefit 

• Gains and ownership (box 11). The sustained influences 
of the intervention after the effects have been assimilated 
within the home settings (e.g., communities, organiza- 
tions, family units) of these beneficiaries 

• Intended outcomes (box 13). Measurable changes in 
behavior of communities, organizations, individuals, 
or systems (e.g., retail sector) resulting from actions of 
these beneficiaries 


The ultimate goal of any intervention is to produce intended 
outcomes and thereby contribute, along with other successful in- 
terventions, to the attainment of desired impacts (box 15) relating 
to the mission (box 1). Cohen and Kibel describe the transitions 
between each of these stages (e.g., from strategic to tactical plan- 
ning) as "gates'' that represent critical points in the implementa- 
tion of an intervention. Effective performance at each gate increases 
the probability for ultimate success. Conversely, lack of planning 
or poor or flawed performance decreases this probability For ex- 





















Figure 1. Stages of Implementation of an Intervention. 

Adapted from Sundberg 0977). 


ample, if the operational planning of the intervention (transition 
point 6) is flawed, then the design may be poorly executed and 
cause the intervention to fail to meet its potential for immediate 
effects, gains, and outcomes. Similarly, if there are few opportuni- 
ties for the effects of the intervention to be assimilated within the 
home context (transition point 10), then the immediate effects of 
the intervention may not translate to gains or outcomes. 

The double arrow icons used to designate the transitions be- 
tween stages signify that these transitions are dialogical, interac- 
tive junctions where the likelihood of future results attributable 
to the intervention can be reduced or, in some cases, increased. 
Hence, rather than be viewed as ''cause-and-effecf' arrows, it is 
useful to interpret these as "communicate-and-create'' arrows. 
Consistent with this view of intervention implementation, sug- 
gested activities for the evaluator to perform include these: 

• Work with others (e.g., stakeholders, designers, beneficia- 
ries) to establish appropriate performance criteria associ- 
ated with successful execution of each aspect of the 
intervention implementation process. 

• Work with others to determine how best to satisfy the 
performance criteria at each gate (i.e., transition point) 

• Monitor actual execution of the intervention using the 
performance criteria as gauges. 

• Establish progress indicators to gauge the degree of 
success at each successive stage in the implementation 
process (e.g., developing clear tactical plans, completing 
all components of the design, attaining immediate effects 
from the execution). 

• Immediately report results from these monitoring activities 
to those involved to permit midstream adjustments in the 
intervention that increase the likelihood of future success 

• Suggest additional, complementary interventions for 

Returning to the earlier point raised regarding evaluator ob- 
jectivity, we argue that objectivity is established and sustained 
through constant attention to agreed-upon performance criteria 
and progress indicators covering all stages of the implementation 


process. The evalua^or helps no one by emphasizing some crite- 
ria or indicators while deemphasizing others or by claiming suc- 
cesses at intermediate stages of the process that are not warranted. 
Instead, the evaluator fulfills the roles of scorekeeper, play-by- 
play analyst, and assistant coach by being ruthlessly objective in 
holding everyone mutually accountable for satisfying all perfor- 
mance criteria. In this way, the evaluator serves a function paral- 
lel to that performed by the quality assurance officer in a 
production facility. In short, the most needed role of the evaluator 
in working with community coalitions is to assume lead respon- 
sibility for quality assurance and quality control during design 
and execution of all interventions. 

A Collaborative Role for the Evaluator 

A critically important shift in responsibility for the evaluator of 
substance abuse prevention activities is being suggested. The 
evaluator must measure outcomes resulting from interventions. 
In addition, the evaluator works with coalition partners to increase 
the likelihood that intended outcomes occur. For this reason, we 
refer to the function as outcome-focused evaluation. That is, evalua- 
tion which is focused on attaining desired outcomes. 

Note that this is not process evaluation in the usual sense. True, 
the emphasis is on the processes leading to outcomes, but the 
evaluator is being challenged to do far more than describe the 
processes. The evaluator, working collaboratively with others con- 
cerned with realizing these outcomes, helps steer the processes in 
the direction of high performance and payoffs. The evaluator does 
not simply react to the designs of others; the evaluator helps shape 
the designs. The evaluator does not simply confirm causality by 
linking interventions to outcomes; the evaluator helps precipi- 
tate the outcomes. 

Working in a multiple-intervention arena, the evaluator charts 
the progress of each component intervention, offers suggestions 
for midcourse adjustments, and forwards (or assesses) proposals 
for additional interventions. Other partners in the coalition de- 
pend on the evaluator to be knowledgeable and candid: they re- 
alize increasingly that the complexity of the undertaking demands 


the type of feedback that the e valuator can provide. More and 
more, they also realize that the absence of this feedback can well 
lead to diminished likelihood of success of the coalition. 

While much of the feedback will result from the expanded, col- 
laborative role for the evaluator as outlined above, some invaluable 
feedback can be derived from more familiar approaches. We next 
discuss this dimension of the evaluation of coalition activities. 

Part II. Bridging to Research-Based 

Testing Hypotheses Related to ttie 
Multiple-Gates Model 

Linkage between the execution of an intervention and its intended 
outcomes (and contributions toward desired impacts) consists of 
a series of stages (Figure 1). This interpretation provides a mecha- 
nism for capitalizing on research methods associated with quasi- 
experimental design models (Campbell & Stanley, 1963; Cook & 
Campbell, 1979). The process of implementing an intervention is 
rightly viewed as a sequence of "mini-experiments'' each suscep- 
tible to study using traditional research tools. For example, the 
evaluator can investigate which aspects of the design parameters 
lead to desired immediate effects. The evaluator might also study 
the factors in the beneficiary's home environment (e.g., commu- 
nity, organization, family) that inhibit or promote the effects to be 
realized as sustained gains. 

The short time segments being studied allow the use of the uni- 
directional cause-and-effect logic that underlies statistical analysis. 
Segmenting the intervention implementation process in this way 
also allows the effects of the intervention to be more easily differ- 
entiated from those stemming from other interventions or exog- 
enous factors. In general, the hypotheses selected to study an 
intervention will depend upon the theory on which the interven- 
tion is founded, the logical assertions made, the critical steps in 
implementing the intervention, and the anticipated effects, gains, 
and outcomes. The challenge facing the evaluator will be to iso- 



late the critical transition points (gates) and apply appropriate 
research strategies to investigate the dynamics at work at each of 
these points. Time, funds, and the scope of the project will dictate 
how extensive or focused this research can and should be. These 
considerations are illustrated in the following hypothetical example. 

A community coalition focused on substance abuse preven- 
tion has strong participation by the local business sector and de- 
cides to reach youths in a high-risk environment through their 
working parents. To accomplish this, they develop a training work- 
shop for working parents that provides useful substance infor- 
mation while emphasizing effective parent-child communication 
and parenting skills. The program consists of 10 modules, each 
which is 1 hour in length. Twelve large businesses agree to spon- 
sor these workshops at their respective work sites, using either a 
'Tunch and learn'' or "work release'' approach. 

At each site, a local program facilitator is selected (usually 
someone with previous training experience) and afforded time to 
attend a course for training facilitators of the 10-module curricu- 
lum. Each facilitator is then assigned the task of promoting the 
program at his or her company, recruiting employees to partici- 
pate, and delivering the training to at least 100 employees during 
the next 12 months. If these targets are met, the collective result 
will be that 1,200 employees will receive the training and, in turn, 
interact effectively with their children (2,400 in number, assum- 
ing 2 children per employee) at home. 

A number of important and researchable questions are raised 
by this intervention. Among these are the following: 

• Are efforts to promote the program successful (i.e., are 
the 100-person targets met at each site)? Are the "right" 
persons attracted (i.e., parents of youths in high-risk 

• Was the facilitator training effective (i.e., did the facilita- 
tors master each of the 10 modules)? What problems were 
encountered in delivering these modules? 

• What new knowledge did the employees gain? What new 
parenting and communication skills did they acquire? 
How can these gains be measured? 


• Did the ''lunch and learn'' or "work release" approach 
prove more effective in (a) initial recruitment, (b) reten- 
tion for all 10 modules, (c) satisfaction with the training, 
and (d) amount learned? 

• What were the outcomes of the program in terms of 
enhanced parent-child interactions? Were there any 
measurable impacts (e.g., improved school performance 
and behavior) that can be attributed to these interactions? 

Continuing with this example, let us consider one research- 
able issue in some detail. Large segments of the general popula- 
tion do not view alcohol or tobacco as drugs. One of the 10 modules 
of the training curriculum focuses on this misconception and in- 
cludes effective ways of discussing the topic at home. It is hy- 
pothesized that the proportion of the 1,200 trainees (and 2,400 
children) who consider alcohol and tobacco to be drugs will rise 
significantly as a consequence of exposure to this module. Ap- 
propriately designed parent and youth pretests and posttests are 
developed to investigate this hypothesis. 

Should early returns from the analysis of pre- and posttests 
suggest that the change in attitude of employees and /or their 
children is minimal, major changes in the design or execution of 
the training might be called for. Irrespective of other environmental 
contributors impinging upon the results, it is reasonable to sug- 
gest that the intervention was not able to effectively negotiate one 
or more gates (e.g., design of the training, delivery, assimilation 
by the trainees, appropriate application at home). Focus groups 
might then be used to isolate potential problem areas that can be 
remedied. Note that the use here of "traditional" tools sparks a 
response that dovetails with the responsibilities of the outcome- 
focused evaluator. 

In general, the appropriate and timely use of "traditional" re- 
search methods nicely supplements the feedback afforded through 
outcome-focused evaluation. It adds to the credibility of the over- 
all effort by permitting clearer isolation of factors contributing to 
(or inhibiting) successful implementation of each intervention. 
Moreover, it permits lessons learned about the efforts under way 
to be more generalizable, and thereby transferable, to other local 
interventions or to prevention programs elsewhere. 



Compensating for the Absence 
of Suitable Controis 

In addition to assessing the factors and processes associated with 
implementation of interventions, the evaluator is challenged to 
gauge the outcomes resulting from these interventions. The evalu- 
ator will rarely have access to adequate control groups to serve as 
references for assessing changes attributable to an intervention. 
However, there are certain strategies that can sometimes be used 
to compensate for the absence of suitable controls. 

Time Series Designs 

Program evaluators with the capacity for performing periodic data 
collection may establish baselines and stable trends to serve as 
reference markers. Statistical techniques such as Box-Jenkins can 
be used to study the compiled time series data and determine if 
the trends following the execution of an intervention are a con- 
tinuation of preintervention trends or a significant departure from 
them. In some cases, data on certain substance abuse indicators 
are already being collected on a frequent and regular basis (e.g., 
hospital emergency room visits linked to abuse). If the interven- 
tion is designed to contribute toward changes in these same indi- 
cators, then time series analysis may be one means for teasing out 
the impact of this specific intervention. 

Individual Growth Curve Models 

Another single system evaluation design (i.e., no control group) 
that can be applied to outcome evaluations is the individual 
growth curve model successfully used in educational research 
(Anderson et al, 1980; Bryk & Raudenbush, 1987, 1988; Bryk, 
Strenio, & Weisberg, 1980; Bryk & Weisberg, 1977; Kim, McLeod, 
& Shantzis, 1990, 1992; Strenio, Weisberg, & Bryk, 1983). In this 
design, the treatment effect of an intervention is determined as 
the difference between the post- and pretest scores that, in turn, is 
adjusted by a natural maturation rate predicted from individual 
growth curve models based on pretest data distribution. 


Within-Experimental Control Design 

It is sometimes possible to generate a " quasi-con trol group." This 
is achieved by isolating a portion of the participants (e.g., the bot- 
tom 2.5 to 5 percent of clients with the lowest level of participa- 
tion). The underlying assumption is that the isolated group is 
essentially similar to the rest of the population in most important 
characteristics, except for the degree to which they were exposed 
to the intervention. Pentz (1993), for example, has reported that 
school-based drug abuse prevention programs with extreme de- 
viations from an established program protocol (i.e., insufficient 
level of program intervention) yield results that are similar to no- 
intervention control groups. 

Meta-Analytic Controls 

An alternative way of estimating the treatment effect in the ab- 
sence of a control group is to estimate expected changes in pre- 
and posttest scores based on a meta-analytic review of existing 
literature, using comparable attitudinal and behavioral measure- 
ments with similar groups of a target population (Kim et al., 1992). 
Specifically, the information of interest would be the tendency of 
similar scale /measurement scores to change over a given period 
for a given population (i.e., whether the scale score changes or 
remains at about the same level between the tested periods). Us- 
ing the average or the general tendency of these scales, one can 
extrapolate the expected changes in the treatment group in the 
absence of the intervention and deduct expected changes from 
the observed score changes between the post- and pretests. 

One-Third S Rule as a Meaningful Change 

A certain level of change may be preselected as constituting the 
minimal amount needed for the intervention to be considered to 
have had a meaningful impact. For example, one is advised to 
apply the one-third s criterion (i.e., a positive change between pre- 
and posttests, whereby the magnitude of change observed is 
greater than one-third of the pooled standard deviation of the 
criterion variable originating from both pre- and posttest 


instruments). The one-third s cutoff is chosen somewhat arbitrarily, 
but it reflects the fact that many social and educational interven- 
tions do not produce dramatic changes in the outcome measure- 
ments within reasonably short periods. In its evaluation of 
educational demonstration projects for the National Diffusion 
Network, the U.S. Department of Education applied this one-third 
s rule in its determination of programs that were considered "edu- 
cationally meaningful.'' That is, a one-third change in the stan- 
dard deviation of the target condition classified a program as 

Multiple Time Series Control Method 

This evaluation strategy combines the techniques of time series 
design with the meta-analytic control method. An interesting ex- 
ample is the evaluation of Connecticut's crackdown on highway 
speeding. After observing not only a reduction in the highway 
fatality rate after the crackdown, but also an unstable up-and- 
down pattern for many years, the evaluator was unsure whether 
the crackdown had any impact. Accordingly, the evaluator com- 
pared statistics from four neighboring States where there were no 
changes in highway traffic enforcement. After observing no simi- 
lar drop in fatality rates in the neighboring States, the evaluator 
was in a better position to claim the ''success" of the project. This 
"comparison lent credence to the conclusion that the crackdown 
had some effect" (Weiss, 1972). 

As a supplement to the simple time series design and the 
nonequivalent control group design, more than one time series 
analysis can be employed. This design uses multiple, simple time 
series analyses applied to the same indicator in different sites. 
This kind of research is usually based on periodically or regularly 
published statistics, such as death rates, infant mortality, traffic 
fatality rates, violent offenses, property offenses, drug offenses, 
public order offenses, or other health statistics originating from 
nearby communities of the target community and from the target 
community itself. A rigorous, systematic comparison with com- 
parable data from nearby communities over a period provides an 


effective weight of validity to the evaluative generalizations made 
by the project evaluator. 

Retrospective Pretest Design 

This method is based on self-reports of behavior /attitude at the 
time of the posttest (e.g., "My attitude toward the use of tobacco 
has changed for the better since I took the training /workshop''). 
Questions are asked during the posttest period only; there is no 
pretest. The treatment effect is estimated on the basis of the dif- 
ference between the posttest score and retrospective pretest scores 
reported by the clients as part of the posttest. According to Rhodes 
and Jason (1987), subjects are more willing to accurately describe 
their substance abuse behavior on the posttest than they are dur- 
ing the pretest period. In other words, the test respondents are 
less inhibited in reporting their true substance involvement at the 
time of the posttest because their denial decreases over the course 
of the intervention. 

In support of retrospective pretest evaluation design, Sullivan, 
Gulielmo, and Lilly (1986) have argued that the clients themselves 
should be interpreting the effects of an intervention because, pri- 
marily, the experience of change is available completely to them 
alone. Furthermore, the usual outcomes that the evaluators are 
seeking to observe and measure often do not materialize during 
the time available to the evaluator. 

Nonequivalent Control Group Design 

In evaluation designs targeting special populations, it may be dif- 
ficult to construct an equivalent control group. However, many 
substance abuse evaluation studies have used nonequivalent con- 
trol groups with some degree of success. For example, in their 
evaluation of a junior high school-based substance abuse preven- 
tion program, McAlister et al. (1980) used a nearby junior high 
school with demographic characteristics similar to those of the 
experimental group. Such an approach should be used only as a 


last resort and only when the results can be bolstered with other 
supporting evidence. 

Design Guidelines in Evaluating 
Prevention Interventions 

As should be apparent from the preceding discussion, the evalu- 
ator must exercise due care, but also ingenuity, in applying tradi- 
tional research approaches to the study of community-based, 
multiple-intervention programs. In this section, we offer some 
guidelines that may prove useful in designing evaluations in these 

Guideline 1: Check the status of pretest differences between 
the experimental groups and the normative population. 

From the initial pool of dependent variables being targeted 
by a particular intervention, all variables should be excluded 
that have a pretest mean scale score that is not meaningfully 
different from that of the general population. If the experi- 
mental group does not differ from the control population at 
pretest with regard to this variable, then it might be argued 
that there is no way in which the intervention can improve 
the experimental group. Such a ceiling is likely to indicate an 
ineffective intervention, when it may well effectively improve 
the experimental group on other criteria. The conceptual jus- 
tification for this exclusion rule is grounded in the assump- 
tion that criteria that do not differentiate the experimental 
groups from the normal population cannot be valid indica- 
tors of the need for intervention. 

Guideline 2: Subjects whose status cannot be improved 
should be excluded. 

Subjects who do not differ from the normative population at 
pretest should be excluded. This exclusionary rule (and the 
two that follow) are offered on the following basis: (1) The 
subjects who have reached apex (i.e., maximum score of the 
theoretical range of the scale used) at the time of the pretest 
have no room to register their gain at the time of the posttest, 
even if it can be assumed that there has been a real improve- 


merit between pre- and posttest. The limitations of the instru- 
ment or the scale embedded in the instrument itself dictate 
the exclusion of these cases. (2) The natural statistical phe- 
nomenon known as the ''regression toward the mean'' also 
dictates exclusion. The latter is considered to be one of the 
important threats to the internal validity involving single sys- 
tem evaluation designs (Cook & Campbell, 1979). Extreme 
values within any data distribution have a mathematical, as 
well as a natural, tendency to regress toward the mean over 
time so that changes at posttest simply reflect statistical ten- 
dency of chance fluctuation of score values rather than inter- 
vention effects. 

Guideline 3: Subjects below 1.96 standard deviations from 
the norm should be excluded. 

Many substance abuse prevention programs are not designed 
for subjects with severe problems. In community-based abuse 
prevention projects, we therefore propose to exclude subjects 
whose pretest scores are below 1.96 standard derivations from 
the norm found in the general population or other appropri- 
ate reference group. In the absence of any norm, we may arbi- 
trarily exclude subjects whose scores on the pretest reside 
below or at the 2.5 percentile (Cohen & Cohen, 1983). 
Guideline 4: Subjects above 1.96 standard deviations from 
the norm should be excluded. 

Following the same logic, subjects should be excluded whose 
pretest scores are above 1.96 standard deviations from the 
norm found in the general population or other appropriate 
reference group. In the absence of any norm, we may exclude 
subjects whose scores on the pretest reside above or at the 
97.5 percentile if it can be assumed that the subjects do not 
belong to a so-called high-risk group. 
Guideline 5: Apply Type I error .10 with Bonferroni rule. 
There is no sanctity involved in the use of Type I decision error 
set at a .05 level. In the substance abuse prevention field, the 
individual interventions are almost certain to have small 
effects. Accordingly, the use of Type I error at .05 level may 
contribute to unnecessary "trashing" of many viable strate- 
gies that have not reached the .05 level but that, nevertheless. 


have had some meaningful ''prevention'' effect (hence in- 
creasing the Type II error). A Type I error set at a .10 level 
may be a reasonable compromise (Catalano et al., 1993). The 
application of Type I error of .10, however, must be adjusted 
by the number of hypotheses being tested for a particular 
evaluation project. Following the suggestion offered by 
Bonferroni (Miller, 1966), we propose the principle of divid- 
ing the overall Type I error into as many equal parts as there 
are hypotheses, and then setting the prehypothesis significance 
criterion accordingly^ 
Guideline 6: Apply one-tailed test. 

Since the direction of the intended changes in outcome vari- 
ables used in substance abuse prevention research is usually 
known from program content and the theoretical basis of the 
program, there is rarely a need to employ two-tailed tests of 
significance for testing hypotheses. For example, in order for 
a given intervention to be thought to be effective, it must im- 
prove the status of the experimental group. We expect the pre- 
test-posttest difference to indicate this improvement and not 
just a change in status. We would not claim that the interven- 
tion was effective if there was a decline in status at posttest. 
With this directional hypothesis, it is not necessary to test both 
tails of the distribution by dividing the Type I error risk rate. 
In fact, incorrectly applying a two-tailed test increases the prob- 
ability of the Type II error. 

Part III. Some Closing Observations 

In opening this paper, we noted that an institutional revolution 
was under way in the substance abuse prevention arenas of the 
Nation. As students of history will attest, revolutions tend to oc- 
cur in clusters. It is not that one revolution triggers the next; rather, 
the conditions that contribute to revolutionary responses are likely 
to coexist in different or geographically separated contexts. It 

^There is always a trade-off between Type I and Type 11 errors. The greater 
the number of hypotheses that are being tested and the more lenient 
the criteria for Type I error, the greater the risk of Type II error 


should therefore not be surprising to note that a "management 
revolution'' is under way in the private sector of the U.S. (and 
international) economy. What is perhaps surprising are the large 
numbers of parallel conditions and responses common to these 
two revolutions. 

The following have been noted by one of the keenest observ- 
ers of the management revolution, Tom Peters (1987): 

• Successful organizations minimize the layers of organiza- 
tional structure and depend instead on more autono- 
mous, self-regulating units for successful product and 
service delivery. 

• Successful organizations are quality conscious, continu- 
ally working to provide more service at lower cost. 

• Control of successful organizations is sustained through 
the energy, excitement, spirit, and hustle that emanate 
from the central leadership and is expressed in the clarity 
of the organization's strategic vision. 

• Successful organizations do not progress incrementally 
They work to make the organization 300 percent more 
effective, not 10 percent more effective. This demands an 
experimental attitude kept realistic by continual attention 
to what customers have to say. 

• Members of successful organizations are provided with 
opportunities and incentives for self-growth, continuous 
learning, limitless involvement, and team-based action. 

Community-based coalitions are positioned to put into prac- 
tice these success principles. Being ad hoc, loosely structured or- 
ganizations, they do not bear the burdens of excessive structure 
and associated restrictive rules of operation. Being community 
focused, they can be expected to be even more responsive to local 
needs and expressions of satisfaction (or dissatisfaction) among 
those being served. In most cases, the severity or persistence of 
substance abuse problem prevention dictates clear statements of 
purpose behind which the coalition can rally and find meaning 
for its work. Coalitions can create and sustain climates oi informed 
experimentation where the best thinking in the substance abuse 
prevention field is applied creatively and strategically. 



Furthermore, coalition and other community members can be 
afforded increasing opportunities for meaningful involvement in 
the coalition's prevention program. 

The outcome-focused evaluator, buttressed with traditional 
research methods, can foster "successful'' coalitions by provid- 
ing important input to (1) the establishment of strategic objec- 
tives, (2) the selection of appropriate interventions, (3) the design 
of the interventions, (4) the execution of the interventions, and 
(5) the gauging of resulting outcomes. The evaluator can develop 
tools for charting and tracking interventions to help identify gaps 
that need to be addressed. The evaluator can help pinpoint roles 
for community members that maximize their participation and 
sense of involvement. The role of the evaluator is to look back at 
what has been accomplished, but also forward at what is cur- 
rently being executed and at what more will be attempted in the 
future. The evaluator can pose many of the important questions 
that the coalition must answer if success is to be attained; and the 
evaluator can help provide information needed to form reliable 

However, as the number of interventions increases, the evalu- 
ator will be confronted with more tasks to perform than are real- 
istically possible. For the coalition to sustain its momentum, 
members of the coalition will need to assume more and more of 
the activities that have heretofore been assigned to the evaluator. 
Here, then, is a final new role for the evaluator: to train others to 
pose the right questions, gather and array the critical informa- 
tion, draw appropriate conclusions, and provide useful feedback. 

If successfully accomplished, evaluation will become inte- 
grated and synchronized with the planning, design, and execu- 
tion of interventions. At this point, the institutional revolution 
under way in the substance abuse prevention arenas will have 
passed an important threshold and have dramatically increased 
the likelihood for sustained success. 



Anderson, S., Auquier, A., Hauck, W.W., Oates, D., Vandaele, W., & Weisberg, 
H.I. (1980). Statistical methods for comparative studies. New York: John Wiley 
& Sons. 

Bryk, A.S., & Raudenbush, S.W. (1987). Application of hierarchical linear 
models to assessing change. Psychological Bulletin, 201(1), 147-158. 

Bryk, A.S., & Raudenbush, S.W. (1988). Toward a more appropriate 
conceptualization of research on school effects: A three-level hierarchical 
linear model. American Journal of Education, 97(1), 65-108. 

Bryk, A.S., Strenio, J.F., & Weisberg, H.I. (1980). A method for estimating 
treatment effects when individuals are growing. /owmfl/ of Educational Statistics, 
5(1), 5-34. 

Bryk, A.S., & Weisberg, H.I. (1977). Use of the nonequivalent control group 
design when subjects are growing. Psychological Bulletin, 84{5), 950-962. 

Campbell, D., & Stanley, J. (1963). Experimental and quasi-experimental designs 
for research. Chicago: Rand McNally 

Catalano, R.E, Hawkins, J.D., Krenz, C, Gilmore, M., Morrison, D., Wells, E., 
& Abbott, R. (1993). Using research to guide culturally sensitive drug abuse 
prevention. Journal of Consulting and Clinical Psychology. 

Cohen, J., & Cohen, R (1983). Applied multiple regression/correlation analysis for the 
behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum Associates Publishers. 

Cook, T.D., & Campbell, D.T. (1979). Quasi-experimentation: Design and analysis 
issues for field setting. Chicago: Rand McNally. 

House, E. (1992, November 4-7). Address to Plenary Session at the American 
Evaluation Association Conference, Seattle, WA. 

Join Together (1992). Who is really fighting the war on drugs? A national study of 
community-based anti-drug and alcohol activity in America. Boston: Boston 
University School of Public Health. 

Kim, S., Crutchfield, C, Williams, C, & Hepler, N. (1994). An innovative and 
unconventional approach to program evaluation in the field of substance 
abuse prevention: A threshold-gating approach using single system evaluation 
designs. Journal of Community Psychology, Monograph Series, CSAP Special 
Issue, 61-78. 

Kim, S., McLeod, J.H., & Shantzis, C. (1990). A short-term outcome evaluation 
of Fs special program: A revisit using the SCAT inventory. Journal of Drug 
Education, 20(2), 127-138. 


Kim, S., McLeod, J.H., & Shantzis, C. (1992). Cultural competence for evaluators 
working with Asian American communities: Some practical considerations. 
In M.H. Orlandi, R.W. Weston, & L.G. Epstein (Eds.), Cultural competence for 
evaluators. DHHS Publication No. (ADM)92-1844. 

McAlister, A. (1980). Pilot Study of Smoking, Alcohol, and Drug Abuse 
Prevention. American Journal of Public Health, 70, 719-721. 

Miller, R.G., Jr. (1966). Simultaneous statistical inference. New York: McGraw- 

Pacific Institute for Research and Evaluation (1992). Evaluation ofOSAP's national 
training system and community technical assistance services: Final evaluation design. 
Contract #277-91-2004. OSAP, Rockville, MD. 

Pentz, M. A. (1993). Benefits of integrating adolescent health promotion strategies 
in school, community, and medical settings. Journal of the American Medical 

Peters, T. (1987). Thriving on chaos. New York: Harper and Row. 

Rhodes, J.E. & Jason, L.A. (1987). The retrospective pre-test: An alternative 
approach in evaluating drug prevention programs. Journal of Drug Education, 
27(4), 345-356. 

Rossi, P. (1992, November 4-7). Address to Plenary Session at the American 
Evaluation Association Conference, Seattle, WA. 

Strenio, J.L.F., Weisberg, H.I. & Bryk, A.S. (1983). Empirical Bayes estimation of 
individual growth curves parameters and their relationship to covariates. 

Biometrics, 39, 71-86. 

Sullivan, A. P., Gulielmo, R., & Lilly, L. (1986). Evaluating prevention and 
intervention procedures. Journal of Drug Education, 26(1), 91-98. 

Sundberg, N.D. (1977). The assessment of person. New York: Prentice-Hall. 

Weiss, C.H. (1972). Evaluation research. Englewood Cliffs, NJ: Prentice-Hall. 



The Cultural Context of 
Epidemiologic Research^ 

Toshiaki Sasao, Ph.D. 


This chapter describes Sasao and Sue's research framework (the 
''cube'' model), in which community researchers working in eth- 
nic/cultural communities can make appropriate decisions on con- 
ceptual and methodological issues from a culturally anchored, 
ecological contextualist perspective. The intent of the model is to 
articulate ethnic-cultural heterogeneity in community research by 
elucidating three meta-methodological issues: (a) definition of an 
ethnic-cultural community, (b) applicability of cross-cultural 
theories and methods to ethnic-cultural community research, and 
(c) geographical or ecological instability of an ethnic-cultural com- 
munity. The model posits that ethnic-cultural community research 
can be conceptualized as a three-dimensional structure that rep- 
resents an interaction among research questions, methods, and 
cultural complexity (referring to the extent to which an ethnic- 
cultural group is defined in a larger ecological context or commu- 
nity at both the individual and collective levels). Future directions 
for research are discussed in terms of the utility and the limita- 
tions of the proposed research model. 

'Some of the materials in this chapter have been adapted from Sasao 
and Sue (1993). 



Despite the increasing recognition of cultural influences on social 
and health behaviors in recent years (Betancourt & Lopez, 1993; 
Landrine & Klonoff, 1992; Shweder & Sullivan, 1993), systematic 
investigations of the epidemiology of substance use and abuse in 
ethnic-cultural communities have been sorely lacking and lim- 
ited (Cheung, 1989; Trimble, Bolek, & Niemcryk, 1992). This is 
unfortunate because numbers and diversity of various ethnic- 
cultural groups in the United States are indeed increasing, given 
high rates of immigration, interethnic or intercultural marriage, 
and intergroup conflicts, especially in large metropolitan areas. 
Substance abuse researchers working with ethnic-cultural com- 
munities will face a multitude of methodological and conceptual 
challenges because they are compelled to work with the "vicissi- 
tudes of putting the etic to work'' in their investigations (Trimble, 
1988). The relatively slow progress of substance abuse research in 
ethnic-cultural communities (Betancourt & Lopez, 1993; Loo, Fong, 
& Iwamasa, 1988; Speer et al., 1992; Trickett, 1990) is further ag- 
gravated by less than optimal efforts to incorporate values of 
ethnic-cultural heterogeneity into epidemiological research. In ad- 
dition, it should be noted that most of the so-called ethnic- 
cultural research on substance use has focused exclusively on 
analyses by ethnic glosses or investigator-defined ethnic catego- 
ries as independent variables, with little or no attention to cul- 
tural elements (Betancourt & Lopez, 1993; Trimble, 1990-1991). 
Above all, we lack guidelines or directions for conducting epide- 
miological research in ethnic-cultural communities (Sasao & Sue, 
1993; Sue, 1991; Vega, 1992). 

Purpose and Overview 

The purpose of this chapter is to provide a conceptual and meth- 
odological framework in which substance abuse researchers work- 
ing in various ethnic-cultural or multiethnic communities can 
make appropriate decisions on conceptual and methodological 
issues from a culturally anchored, ecological context perspective. 
To accomplish this broad goal, the chapter begins with the defini- 


tion of ethnic-cultural diversity as we conduct community-based 
research. This brief discussion will be followed by a critical ap- 
praisal of problems or common assumptions often made in eth- 
nic-cultural community research. Among other things, these 
problems are related to (1) the arbitrary definition of an ethnic- 
cultural community leading to the paucity of a ''community" fo- 
cus in research, (2) the questionable application of cross-cultural 
methods and concepts to research in ethnic-cultural communi- 
ties, and (3) the geographical or ecological instability of an ethnic- 
cultural community across time. Then, a framework for conducting 
research in ethnic-cultural communities (''cube'' model) will be 
described in which an examination of culturally anchored eco- 
logical contexts is stressed by introducing the concept of cultural 
complexity. Finally, based upon the proposed model, several direc- 
tions for future research as well as the limitations of the model 
will be discussed. 

Understanding Ethnic-Cultural 
Diversity: A Double-Tiered 

What do we mean by "ethnic-cultural diversity"? At its simplest 
level, the concept of such diversity can be defined as a double- 
tiered social phenomenon. First, the within-group diversity refers 
to heterogeneity due to changing or diversifying patterns of social 
attributes or relations in the family, the neighborhood, and /or local 
communities within ethnic-cultural groups including broadly 
defined groups or subgroups (e.g., African Americans, Caribbean 
Blacks; Hispanic /Latino Americans, Mexican Americans, Puerto 
Ricans, Central Americans; Asian Americans, Korean Americans, 
Japanese Americans; and American Indians /Alaskan Natives, 
Navajo Indians [Office for Substance Abuse Prevention, 1991]). 
The within-group diversity is further compounded by other social 
boundaries or orientations such as gay /lesbian issues, women's 
issues, and political or religious orientations. Also, there 
are increasing numbers of intergroup (e.g., racial, ethnic, reli- 
gious) marriages within many ethnic-cultural communities, a 


phenomenon that influences many social interactions across 
groups as well. As noted earlier, generational differences in vari- 
ous ethnic-cultural communities tend to exacerbate the level of 
heterogeneity within a particular group. Even among non- 
Hispanic /Latino Whites, sociologist Richard Alba (1990) argues, 
ethnicity has regained interest because, with the demise of many 
communist countries and changing world politics, U.S. ethnics 
began to realize the appropriateness and relevance of national- 
ethnic identity (e.g., German, Irish, Scots). Therefore, the within- 
group diversity can be conceptualized either at the micro- or 
mesosystem level, while being further influenced by the higher 
order exo- or macrosystem level. 

Second, the other type of diversity is conceptualized as the 
across- or between-group diversity, a reflection of increasing com- 
plexities brought about by growing intergroup tensions between 
various social groups in work, school, or everyday situations (Lam- 
bert & Taylor, 1990; Los Angeles Times, 1992). Although such diver- 
sity situations are explicitly more evident in major urban areas, 
the across-group diversity phenomenon has become of increas- 
ing concern on many college campuses and work settings where 
demographic and social changes have been rather slow until re- 
cently. Policy changes due to such across-group diversity have 
become apparent in student admission or affirmative action em- 
ployment procedures. For example, the population trends in the 
United States, when examined from 1980 to 1990, show an obvi- 
ous example of across-group heterogeneity. There are phenom- 
enal growth rates particularly among ethnic /racial populations 
(e.g., 127 percent increase for Asian and Pacific Islanders, 69.2 per- 
cent increase for individuals from non-Hispanic /Latino origin) 
and a 66.7 percent increase in the "other race'' category An in- 
crease in that category implies increasing within-group diversity 
simultaneously because there is a substantial number of mixed- 
heritage individuals across the United States who cannot identify 
themselves with any of the census-based ethnic-cultural catego- 
ries (Root, 1992). 

The importance of ethnic-cultural diversity or heterogeneity 
for ethnic-cultural community research is that it creates a serious 


dilemma for those who conduct research in ethnic-cultural com- 
munities (Sue, 1991; Zane & Sue, 1986). Community researchers 
working with ethnic-cultural groups are often forced into the de- 
fensive position of having to demonstrate that the etic model does 
not constitute a universal and that cultural differences (the emic) 
do make a difference. This itself becomes a problem because 
ethnic-cultural issues are, by their very nature, paradoxical 
(Rappaport, 1981). Such issues involve two equally valid but con- 
tradictory viewpoints, one emphasizing the importance of differ- 
ences between cultures and the other stressing the significance of 
their commonalities. Consequently, it is important to not become 
too one-sided; otherwise, one perspective dominates to the detri- 
ment of the other. 

In another important way, ethnic-cultural heterogeneity chal- 
lenges the traditional notion of a community, as discussed earlier. 
It is not entirely clear when and how individuals who are exter- 
nally or physiognomically defined members of a certain ethnic- 
cultural group perceive themselves to be part of that community 
Because there are few ''communities'' whose members are ethni- 
cally or culturally homogeneous, "true" community research must 
define an ethnic-cultural community according to criteria other 
than racial /ethnic /cultural categories per se, as well as include 
criteria such as those proposed by McMillan and Chavis (1986): 
perceived membership, a sense of influence, integration and ful- 
fillment of needs, and shared emotional connection. For mem- 
bers of ethnic-cultural groups in the United States, we should note 
the possibility that some of these individuals can fluctuate in their 
actual and perceived group membership, depending on various 
developmental stages and /or diverse behavior settings, such as 
school classrooms or work situations. Thus, the fluid nature of an 
ethnic-cultural community needs to be recognized. In a 
multicultural society such as that of the United States, research 
issues pertinent to ethnic-cultural communities need to be ad- 
dressed in a way that balances solutions to the paradoxes of vari- 
ous ethnic-cultural and mainstream perspectives within and across 
groups in a fluid ecological context or community. 


Assumptions in Ethnic-Culturai 
Community Reseorcln 

Undoubtedly, given the lack of methodological and conceptual 
guidelines for conducting ethnic-cultural community research 
(Milburn, Gary, Booth, & Brown, 1991), it is important to focus on 
improving or innovating existing methods that match research 
questions at appropriate levels of conceptualization for ethnic- 
cultural communities to the extent that both external and internal 
validity would be enhanced (Cook & Campbell, 1979; Seidman, 
1988; Shinn, 1990). Nonetheless, above and beyond various meth- 
odological issues to be resolved (e.g., gaining community entry, 
procuring representative samples, reducing measurement bias, 
increasing respondent compliance), more fundamental meta- 
methodological issues have not been given adequate attention in 
past research efforts involving ethnic-cultural communities. In 
other words, because of our traditional research focus on logical 
positivism, an epistemological perspective that entails the natu- 
ral-scientific canons of reductionism, experimentation, explana- 
tion, operationalization, quantification, and objectivity (Barzun 
& Graff, 1985), community researchers have been shortsighted in 
the understanding and articulation of ethnic-cultural diversity and 
its implications for epidemiological research while failing to meet 
the needs and concerns of local ethnic-cultural communities. For 
example, there has been no clear-cut discussion on what consti- 
tutes an "ethnic-cultural community'' Is it a geographical or rela- 
tional entity? Provided that ethnic-cultural individuals or groups 
are socially embedded within a larger societal context (Szapocznik 
& Kurtines, 1993), it is imperative to investigate the entire eco- 
logical context where a target ethnic-cultural group (e.g., Cuban 
Americans) and other relevant ethnic-cultural groups (e.g., 
Korean Americans) coexist. At least three interrelated meta- 
methodological assumptions require immediate attention as eth- 
nic-cultural community research progresses. 


Defining an Etinnic-Cultural 

One prevalent assumption in research with ethnic-cultural com- 
munities concerns a definition of ethnic-cultural community as it 
is being used by contemporary researchers. Whereas defining 
boundaries of a community, be it geographical or relational, has 
been a major issue of interest within community psychology it- 
self (Chavis & Newbrough, 1986; Chavis, Stucky, & Wanderman, 
1983; Newbrough, 1992), the concept of a community, as used in 
much of the research involving ethnic-cultural groups, has often 
been equated with a group of individuals (research participants) 
who possess certain ethnic-cultural markers (e.g., language, skin 
color, distinct cultural practices). Furthermore, the concept has 
rarely implied a direct reflection of social, historical, and cultural 
experiences and values of such ethnic-cultural individuals in com- 
munity contexts (Evans & Whitfield, 1988; Hall, Evans, & Selice, 
1989; Leong & Whitfield, 1992; Olmedo & Walker, 1990). 

The majority of community-based ethnic-cultural studies that 
have appeared in the empirical literature usually address sub- 
stantive issues (e.g., depression, health status, substance abuse, 
gang violence) by simply classifying ethnic-cultural populations 
into broad "ethnic glosses'' (e.g., Asian Americans, Hispanics), 
thus providing research on "ethnic or cultural differences." Even 
when specific ethnic subgroups (e.g., Caribbean Hispanics) can 
be identified and studied with sample sizes large enough for ac- 
ceptable statistical power, research findings are often limited in 
generalizability to other contexts because large numbers of par- 
ticipants are often recruited, via systematic or captive sampling 
in such intact ethnic groups as churches, professional associations, 
or ethnic studies classes on college campuses. In order to obtain 
larger sample sizes for research, the selection criteria are usually 
based on race or physiognomic factors per se, but there has not 
been any reference to ecological contexts where the participants 
reside, work, or study. Therefore, past research has not adequately 
defined an ethnic-cultural community in a way that captures ex- 
periences of ethnic-cultural individuals in context. 


With the growing promise of an ecological-contextualist epis- 
temology in community research (Berger & Luckman, 1966; Felner, 
Phillips, DuBois, & Lease, 1991; Kingry-Westergaard & Kelly 1990; 
Tolan, Keys, Chertok, & Jason, 1990; Trickett, 1990; Watts, 1992), 
substance abuse researchers must begin defining an ethnic- 
cultural community within a larger societal context in relation to 
other relevant ethnic-cultural categories or groups. This is true 
even when studying a single ethnic-cultural group or category 
Otherwise, ethnic-cultural community research is likely to per- 
petuate an illusion that any ethnic-cultural group automatically 
forms a community itself by virtue of its ascribed or assumed eth- 
nic or cultural attributes. For example, it is tacitly assumed that 
once individuals can be identified as belonging to a certain ethnic- 
cultural group, they share a common understanding of their own 
ethnicity or culture and identification with the group. There is 
mounting evidence to suggest, however, that the population homo- 
geneity assumption in any ethnic-cultural group may not be valid 
(Trimble, 1988). Members within a particular group exhibit con- 
siderable individual differences on a number of variables, such as 
acculturation, language skills, generational status, immigration 
or refugee history, and ethnic self-identification (Phinney, 1991). 
For instance, a third-generation Mexican American is likely to be 
very different from a first-generation Mexican immigrant, not sim- 
ply due to developmental differences in socialization, but also 
because of current developments in interethnic relations in the 
United States and /or changes in global or international political 
situations. In addition, a U.S. -born Mexican individual may also 
be a member of many different "communities,'' including ethnic- 
national, religious-spiritual, occupational, recreational, and po- 
litical. Therefore, defining a community entails more than 
classification based on simple ethnic glosses and requires a closer 
examination of multiple social categories relevant to the individual 
as well as the community 

As another example, in a recent high school drug use survey 
conducted in predominantly Chinese, Vietnamese, and Mexican 
multicultural communities of southern California (Sasao, 1992a), 
approximately 20 percent of the Chinese students indicated that 
their primary cultural identification was Mexican. Although the 


self-perceived ethnicity of these Chinese youths was Chinese, their 
cultural identification was Mexican because these youths lived 
and played in the predominantly Mexican- American community. 
Moreover, a pattern of such cultural identification was significantly 
associated with the perception of the school campus' interracial 
climate. The Chinese students whose cultural identification was 
more Mexican than Chinese felt that the campus climate was more 
congenial than those who felt otherwise. Subsequent face-to-face 
interviews with selected students of Chinese or Mexican back- 
ground revealed that they both shared a great deal of similarities, 
such as immigration status and difficulties in mixing with the 
general student population. Therefore, an ethnic-cultural commu- 
nity must be viewed more as a social-cognitive-cultural-histori- 
cal-contextual entity than as the one based on physiognomic 
attributes and /or geographical boundaries per se (Szapocznik & 
Kurtines, 1993). It is often the case that an ethnic-cultural com- 
munity encompasses more than a single dimension of commu- 
nity attributes, such as geographical, spiritual, or relational 
attributes (Liu, 1980). This type of definition is consistent with 
the current thinking of a theory of community in the postmodern 
world's community psychology (Newbrough, 1992). By defining 
an ethnic-cultural community in a larger ecological-contextual 
framework, community researchers should be able to address 
"true'' community phenomena of interest that focus on an inter- 
action among significant social units or entities (Seidman, 1988), 
such as substance abuse or interethnic climate in work or school 
settings (Green, Adams, & Turner, 1988; Sasao, 1992a). 

Applicability of Cross-Cultural 
Research Methods and Concepts to 
Ethnic-Cultural Community Research 

A further related, but uncontested, assumption in ethnic-cultural 
community research is the applicability of cross-cultural (or cross- 
national) theories and methods to research in ethnic-cultural com- 
munities (Berry, Poortinga, Segall, & Dasen, 1992; Brislin, Lonner, 
& Thorndike, 1973; Triandis, 1992; Triandis & Lambert, 1980). 


Oftentimes, studies comparing different ethnic-cultural groups in 
the United States are not equivalent to studies involving the com- 
parisons between different national groups. Social groups, includ- 
ing ethnic-cultural groups in the United States, often share the 
same ecological contexts or settings (e.g., schools, neighborhoods, 
public facilities) where they do interact with each other by default. 
Comparisons between these groups must take into account not 
only intracultural or intercultural differences, but also differences 
in the dynamics of social interactions between the groups (e.g., 
the experience of prejudice and discrimination) in ecological con- 
texts or settings. For example, comparisons between Chinese in 
China and Whites in the United States are not comparable to those 
between Chinese Americans and Whites in this country. Although 
Chinese Americans may well maintain some of the cultural val- 
ues seen in China, they have also had years of interacting with 
Whites and other Americans of different backgrounds. Moreover, 
another layer of complexity can be added to the experience of 
Chinese Americans that includes animosities based on current and 
past historical events between China and the United States, which 
also influences the ''texture'' of the Chinese-American commu- 
nity. This interaction means that while cross-cultural research can 
study different groups as independent variables, ethnic-cultural 
groups in the same context interact and are not independent; thus, 
in investigating issues of concern in ethnic-cultural communities 
(e.g., substance abuse or violence), research methods must accom- 
modate both historical and international relations, as well as geo- 
graphical and relational components. 

In addition, some methodological issues are applicable to both 
cross-cultural psychology research and ethnic-cultural commu- 
nity research. They are typically expressed in terms of the etic- 
emic distinction (Sue, 1991), discussed later in this chapter, which 
involves such issues as adequacy of sampling procedures, cul- 
tural response sets, and conceptual equivalence of measures for 
different groups. However, unlike cross-cultural research, ethnic- 
cultural community research has been conducted in an increas- 
ingly larger and ethnically diverse societal context, where more 
than one, single ethnic-cultural community usually coexists. Thus, 
analytic methods or strategies for investigating phenomena rel- 


evant to ethnic-cultural communities must be tempered by the 
ecological concerns for these communities, such as immediate 
neighborhoods and school climate. Unfortunately, although the 
contextualist perspective allows an examination of ecological- 
contextual factors in community psychology research, there is no 
methodological provision for incorporating the notion of an 
''ethnic-cultural group in context'' into community research. Such 
a provision would provide guiding research principles that inte- 
grate ethnicity /culture-specific issues, contextual issues, and the 
dynamics of community infrastructures: for instance, local com- 
munity leaders and community residents (Milburn et al, 1991; 
Watts, 1992). 

Previous methodological research efforts have focused exclu- 
sively on improving psychometric properties of measures for 
ethnic-cultural groups, ensuring conceptual equivalence across 
different groups, procuring adequate representative samples, find- 
ing appropriate control or comparison groups in intervention- 
oriented research, enhancing community support for scientifically 
viable research, gaining community entry, selecting qualitative 
versus quantitative methods, deciding on information dissemi- 
nation methods, and improving academic researcher-community 
relations (Cervantes & Acosta, 1992; Liu, 1980; Marin & Marin, 
1991; Milburn et al, 1991; Sue & Morishima, 1982). However, even 
though these issues have been discussed with respect to popula- 
tion-specific groups identified by broad ethnic categories such as 
African Americans, Asian Americans, and Hispanic /Latino Ameri- 
cans (Marin & Marin, 1991; Rogler, 1989; Sue & Morishima, 1982), 
the growing diversity and heterogeneity of each population re- 
quires these population-specific methodological issues to be con- 
sidered in the larger societal context or ecological setting. For 
instance, there have been hardly any methodological advances in 
addressing issues and defining a community with mixed heri- 
tage individuals such as Amerasians, the growing, but largely 
hidden and ignored population in many parts of the United States 
(Root, 1992). Many basic questions remain unanswered, includ- 
ing whether it is appropriate or possible to define a multicultural 
or mixed race community for research, or whether a separate com- 
munity-based research for this often neglected group is warranted. 


Geographical or Ecological 
Instability of an Ethnic-Cultural 

A third problem or assumption in translating diversity issues into 
community research is that once a target community is identified 
in a larger ecological context or community, where it is embed- 
ded along with other groups, the ethnic-cultural community re- 
mains stable from one such community to another (e.g., an 
African-American community in Los Angeles is comparable to 
another in Detroit or Baltimore). However, there is enough evi- 
dence to show that ecological factors make huge differences when 
explaining certain community phenomena from one location to 
another. For example, alcohol researchers are cautious in general- 
izing research findings on alcohol use among Japanese Ameri- 
cans in Hawaii to other Japanese Americans in the mainland 
United States because of different ecological contexts, such as 
interethnic relations (Johnson & Nagoshi, 1989). In the past, 
community-based research has generated controversy and con- 
sternation because of divergent views on research issues and 
methods as represented by the researcher and the local commu- 
nity. Although this concern has led to the development of em- 
powerment and sense-of-ownership notions in community 
research (Rappaport, 1987), it has been restricted to promoting 
empowerment w/f/im an ethnic-cultural group (e.g., empowerment 
in the Mexican- American community in East Los Angeles), but it 
has rarely extended to across groups or multicultural groups 
within a larger ecological context or setting. Thus, community 
research needs to be conducted with due consideration given to 
the intricate community process that represents different episte- 
mological orientations of constituents, including community resi- 
dents, leaders, and politicians of a target ethnic-cultural group 
and other relevant groups in a specific ecological context or com- 
munity — for example, Rappaport's ''paradoxes in community re- 
search'' (1987). Also, particularly important for ethnic-cultural 
communities are generational differences within an ethnic or cul- 
tural group. For example, while the local community leadership 


in the Japanese- American community is primarily held by sec- 
ond or later generations of Japanese Americans {nisei or sansei), 
the business leadership in the same community is often controlled 
by the first-generation, Japan-born individuals (shin-issei). Such 
generational differences may lead to a different set of local com- 
munity needs, thereby resulting in divergent views or conflicts 
about community priorities. 

The foregoing discussion indicates that research in ethnic- 
cultural communities is often fraught with methodological and 
conceptual issues. These issues not only stem from validity and 
reliability concerns from the traditional research standpoint, but 
also are oitenmeta-methodological in nature. The meta-methodologi- 
cal issues are concerned with relative inattention to ecological 
contexts where demographics are constantly and rapidly chang- 
ing, as well as relative negligence of complex sociopolitical cli- 
mates in which various stakeholders represent ethnic-cultural 
communities. This further suggests that basic concepts in com- 
munity-based research, such as community and race-ethnicity- 
culture, need to be reexamined in order to guide and promote 
future research in ethnic-cultural communities. This research 
should not dwell on such surface or external social categories as 
skin color or language; however, it should focus on con\munity 
phenomena represented by a transaction of various units or enti- 
ties that constitutes a social system or setting, or what Seidman 
(1988, p. 93) calls social regularities, which are 'Targer than the family 
or between settings or systems'' (Bronfenbrenner, 1979; Szapocznik 
& Kurtines, 1993). Instead of focusing research exclusively on 
microsystem levels (such as dyadic or marital relations) in ethnic- 
cultural "community'' research, it is important to generate social 
regularities (as opposed to individual-based phenomena) for 
ethnic-cultural or multicultural communities. These regularities 
can lead to a body of knowledge useful for culturally anchored 
"community" prevention and treatment interventions. In addi- 
tion, a trade-off between methodological and substantive signifi- 
cance of research must be negotiated or balanced with such 
concerns as ethnic-cultural diversity and the needs associated with 
local communities. 



Toward a Culturally Anchored 
Ecological Framework of Research 
in Ethnic-Cultural Communities 

Ethnic-cultural community research calls for a research framework 
that provides an integration of methodology and conceptualiza- 
tion that addresses questions relevant to each ethnic-cultural com- 
munity. It is important that such a community be understood in 
the larger context of the society where the community of interest 
is defined and embedded. 

The "Cube" Model 

Sasao and Sue (1993) proposed a research model in order to ar- 
ticulate ethnic-cultural diversity by incorporating the concept of 
cultural complexity, based on social psychological theories of group 
behaviors and beliefs: social identity theory (Tajfel & Turner, 1986); 
self-categorization theory (Turner, Hogg, Oakes, Reicher, & 
Wetherell, 1987); and "social representations'' (Moscovici, 1984) 
into the study of community phenomena relevant to ethnic-cultural 
groups in contexts. More specifically, the model attempts to eluci- 
date issues related to three meta-methodological issues discussed 
earlier: (1) definition of an ethnic-cultural community, (2) appli- 
cability of cross-cultural theories and methods to ethnic-cultural 
community research, and (3) geographical or ecological instabil- 
ity of an ethnic-cultural community. As shown in Figure 1, ethnic- 
cultural community research can be conceptualized as a 
three-dimensional figure that represents a transaction among types 
of research questions being asked, selection of methods, and cul- 
tural complexity (referring to the extent to which an ethnic- 
cultural group is defined in a larger ecological context or commu- 
nity, at both the individual and collective levels). It is argued that 
in designing and conducting research in ethnic-cultural commu- 
nities, these three elements will interact to determine the design 
of a study as well as outcomes; therefore, they must be examined 
simultaneously and weighed against one another to obtain scien- 
tifically valid research, despite constraints due to increasing di- 
versity, as discussed earlier. 





("Ethnic Gloss") 

^ (Needs Assessment) 


(Program Evaluation) 

Quantitative Qualitative 

Surveys Clinical Observation 

Single Subjects Ethnographic Observation 

Archival Data Focus Group Interview 


Figure 1 . Toward a Culturally Anchored Ecological Model of 
Research in Ethnic-Cultural Communities. 

Source: Sasao & Sue (1993) 


Research Questions 

Given the needs of local ethnic-cultural communities and research- 
ers' interests, three types of research questions are typically gen- 
erated. Descriptive or epidemiological questions are the first step in 
any community-based research, usually in the context of commu- 
nity needs assessment. These questions include the following: 
What is the prevalence of substance abuse among Asian Pacific 
Americans or African-American communities (Sasao, 1992b)? 
What kind of mental health needs exist in the low-income His- 
panic/Latino community? What patterns of high-risk sexual ac- 
tivities exist among young urban African- American males? Not 
only are these questions of interest to community researchers, but 
they also attract the attention of many community-based organi- 
zations for local and Federal lobbying and funding purposes. 

In addition to epidemiological information, community re- 
search also focuses onetiological or explanatory issues. Typical ques- 
tions include these: What is the etiology of substance abuse and 
early teenage pregnancy among young African- American females 
living in public housing? Is the lack of life skills a major explana- 
tion for poor academic achievement and low self-esteem among 
Cambodian refugee children and parents (Botvin, 1986; Schinke, 
Botvin, & Orlandi, 1992)? Are theories of alcohol expectancy ap- 
plicable to both males and females in the Korean- American com- 
munity (Goldman, Brown, & Christiansen, 1987). Another question 
might be what some risk factors are for drug abuse for urban 
African- American adolescents (Farrell, Danish, & Howard, 1992). 
In one study, theoretical models were developed to explain social 
integration and social support among Mexican Americans and non- 
Hispanic /Latino Whites (Golding & Baezconde-Garbanati, 1990). 

Finally, prevention/treatment intervention questions particularly 
are of greater importance to community-based service providers 
and program evaluators, as well as to Federal or other funding 
agencies, especially in view of a tightening economy. For example, 
the effectiveness of a street-based AIDS prevention and health 
education program for intravenous drug users in San Francisco's 
multiethnic communities was investigated (Watters et al., 1990). 
Another question is what the most effective prevention or treat- 


ment modalities for new immigrant youth (e.g., life skills train- 
ing, stress management, peer support, detoxification, residential 
programs) would be. 

In order to approach some or a combination of these ques- 
tions, the next obvious step concerns the choice of methods for 


For ethnic-cultural community researchers, a range of traditional 
quantitative and qualitative methods (shown in Figure 1) is avail- 
able in answering the above questions. Although these methods 
are familiar to social scientists, the applicability of these methods 
in ethnic-cultural communities needs to be determined based on 
multiple perspectives on research and the type of research ques- 
tions asked in a specific ecological context or setting. For instance, 
although such qualitative methods as ethnographic observation 
and focus group methods are often viewed with some skepticism 
by social scientists because of the lack of systematic standards in 
reliability and validity, certain contexts in which ethnic-cultural 
groups reside or work call for these qualitative methods because 
other quantitative methods are perceived as being "not culturally 
appropriate," or because local communities tend to distrust quan- 
titative methodology, arguing that such research has not taken 
historical or global considerations of the communities into account. 
Therefore, merely knowing what questions are being asked and 
which methods are available and appropriate are not sufficient to 
conduct community research in ethnic-cultural groups. Notwith- 
standing that efforts to integrate and systematize qualitative meth- 
ods are encouraged (Miles & Huberman, 1984) so that they may 
become amenable to quantitative analysis, it can be argued that 
the choice of methods in certain ecological contexts is often con- 
tingent on the nature of ecological contexts as well as perspec- 
tives represented in such contexts. 

Cultural Complexity 

In this model, the concept of cultural complexity is introduced 
to allow the identification and assessment of community 


phenomena or social regularities in culturally anchored ecologi- 
cal settings (Seidman, 1990). The concept of cultural complexity 
can be defined at two levels. First, it is defined, at the individual 
level, as the degree to which an individual is defined not only by 
his or her racial-ethnic-cultural category, but also by his or her 
own affective, behavioral, and cognitive representation of that 
social category — which is defined as one's social identity in the 
social psychological literature (Taj f el & Turner, 1986). Second, it is 
conceptualized at the larger, collective level of a context or setting 
(e.g., community, neighborhood, school) where individual mem- 
bers are located or embedded. In fact, past community psychol- 
ogy research should have concerned itself with this 
extra-individual level of community phenomena (Pel ton & Shinn, 
1992; Seidman, 1990); however, most ethnic-cultural "community" 
studies found in the literature have focused exclusively on per- 
sonality or motivation, and interpersonal or family dynamics 
(Snowden, 1987). At the collective (as opposed to individual level), 
the concept of cultural complexity is defined as the extent to which 
a relevant group entity, such as a predominantly Chinese com- 
munity or neighborhood, Hispanic /Latino gay/ lesbian commu- 
nity, is defined by itself or others vis-a-vis other existing relevant 
social categories (e.g., an ethnic Vietnamese-Chinese community) 
within the same ecological setting. The concept here is closely 
related to social representations in social psychology (Moscovici, 
1984) or representations in classical sociology (Durkheim, 1898) 
because cultural complexity at this level refers to the actual or 
perceived degree of identification as a group or community at the 
macrosystem's level (Bronfenbrenner, 1979), not at the individual 
or microsystem's level. 

In Figure 1, depending upon complexity at these two levels, 
three layers of cultural complexity can be examined. The first level 
of cultural complexity can be most appropriately described as 
acultural complexity because researchers collect data based on such 
physical markers as ethnic glosses or physical characteristics, with- 
out regard to the ecological context of the research setting, and 
analyze data using ethnicity /culture as a categorical variable. Also, 
the individual members' social identity is blatantly ignored in 


collecting and interpreting such data. Unfortunately, the majority 
of the so-called ethnic-cultural community research falls into this 
category In some cases in which this type of ethnic-cultural com- 
munity research is inevitable, especially when examining archi- 
val data collected by other investigators, an extensive analysis 
must be performed in which both mediating and moderating vari- 
ables relevant to identified culturally anchored social contexts (e.g., 
socioeconomic status, crime rates in neighborhoods obtained from 
the police) are examined by additional data collection and analysis. 

The second layer of cultural complexity is ethnocultural com- 
plexity, in which the community or group being studied must be 
defined by members of the community or group, not only by the 
ethnic group of interest, but by members of other ethnic-cultural 
groups, as well, within the same ecological context. In this clearly 
identified community, community researchers are required to as- 
sess the community members' perception of group identification 
and cohesion vis-a-vis other relevant social categories in order to 
establish the definition of a target community. This notion of cul- 
tural complexity is perhaps most common in community research 
with ethnic-cultural groups; however, the focus is usually on one 
single population, such as Hispanic /Latino residents in inner-city 
public housing. 

In research involving illegal or hidden populations within 
certain ethnic-cultural groups, such as youth gangs or drug abus- 
ers, the next layer of cultural complexity in Figure 1 is important. 
In the subcultural context, individual members are no longer de- 
fined according to imposed social categories such as race /ethnicity 
per se, but sources for the definition of their own subculture and 
individual members' social identity in that subculture or "street'' 
culture stem from a combination of various cultural elements or 
categories. For instance, in a Vietnamese gang subculture, heavy 
substance abuse has been observed by community workers (Sasao, 
1992b). Factors related to more than one single culture (i.e., Viet- 
namese) must be examined in this setting because Vietnamese 
youths were found to be involved with the Hispanic /Latino or 
Filipino youth gangs initially, and, therefore, an explanation for 
Vietnamese gang-related and drug-related behaviors can be best 


understood as a combination of multiple cultural elements en- 
meshed at the street, subcultural context (e.g., Vietnamese family 
structures, cultural elements learned from Hispanic /Latino or Fili- 
pino gang activities, interethnic discrimination experiences). 

Thus, the model presented implies that in order to identify 
and address culturally anchored social regularities, a community 
needs to be defined by more clearly incorporating the concept of 
cultural complexity in the research design, such as self-perceived 
identification with an ethnic /racial category or acculturation sta- 
tus (Trimble, 1988, 1990-1991), or how a community or neighbor- 
hood is viewed or defined in the context of other relevant social 
categories. In defining an ethnic-cultural community for research, 
we need to go one step beyond external or imposed definition of 
individuals or groups based on physical characteristics, such as 
skin color, physical attributes, or geographical dispersion, and 
view the concept of community as a social-cognitive-contextual 
entity The real task for community researchers would be to inte- 
grate data that stem from both levels of cultural complexity, that 
is, the individual and the community, in identifying an ethnic- 
cultural community. 

It is also important to note that decisions about which ques- 
tion is answered by what methods and in what contexts are de- 
termined by multiple perspectives represented in a particular 
research project. In each of these different cultural-complexity 
contexts, some methods are more feasible and preferable than oth- 
ers. Although a carefully conducted sample survey (a quantita- 
tive approach) may yield greater generalizability to other settings, 
survey techniques could be inappropriate to use with youth gang 
members in the subcultural, street ecological context. In sampling 
"hidden and rare'' populations, such as the homeless mentally ill 
(Koegel, Burnam, & Farr, 1986) or intravenous drug abusers liv- 
ing on skid rows, such appropriate qualitative methods as a key 
informant survey, a focus group, or participant observation may 
be used to obtain proxy "epidemiologicar' data. Although these 
methods do not provide entirely accurate estimates, they are help- 
ful in generating initial exploratory hypotheses about these popu- 
lations, and also are complementary to more traditional 
quantitative methods. 


By examining how three dimensions (type of research ques- 
tions, methods, and cultural complexity) transact with each other, 
potential solutions to some of the overarching issues discussed 
previously may be obtained. For instance, most of the claims or 
social stereotypes or sources of misunderstanding regarding the 
low substance abuse among Asians (Zane & Sasao, 1992) appear 
to be based on either inadequate methodology or inappropriate 
cultural contexts where research was conducted, or a combina- 
tion of both. Because Asian substance abusers are usually hidden 
in Asian ethnic communities (Sasao, 1992b), using a traditional 
survey method in an ethnic-cultural community would not pro- 
vide an accurate estimate of alcohol and drug abuse in a certain 
Asian community For example, we might begin identifying and 
defining certain social regularities or contexts in which Asians 
find themselves likely to use alcohol or drugs (e.g., when alone, 
when with other male friends, or on cultural festive occasions), 
and then conduct an extensive ethnographic study to generate 
h3rpotheses prior to investigating epidemiological questions. 

Furthermore, the use of multiple methods is desired, when- 
ever possible, because using multiple methods not only leads to 
validation of findings, but also pushes for ecological validation 
as an ongoing process of testing assumptions about a certain phe- 
nomenon in different culturally anchored ecological contexts of 
research. Similarly, Trickett, Watts, and Birman (1992) argue that 
an ecological approach needs to serve as a heuristic as we inte- 
grate diversity issues into community psychology research, and 
also that the reemergence of the social constructionist and 
contextualist philosophy of science, as an alternative to logical 
positivism, can be seen as a societal recognition that ethnic-cul- 
tural diversity has become an issue that must be incorporated into 
our community-based research. 

Implications for Future Researcti 

The model presented in this chapter suggests some courses of 
action for future research in ethnic-cultural commiunities, includ- 
ing the following: 


In conceptualizing and designing research projects, 
researchers working in ethnic-cultural communities 
should consciously be guided by a consideration of 
interrelated factors, such as the ones we have identified in 
the ''cube'' model involving cultural complexity, method- 
ology, and type of questions being addressed. The model 
also provides some directions in the course of research. 
For example, in the case of ethnic-cultural groups in 
which little research has been conducted, it may be wise 
for investigators to initially focus on descriptive ques- 
tions, use more qualitative methods, and examine the 
groups using low levels of cultural complexity. Such 
research may help to establish a baseline of knowledge, 
identify relevant parameters and variables prior to more 
intensive study, and gain insight into the appropriate 
methodologies and instruments to use. 
Future investigations of a target ethnic-cultural group 
should begin collaborating with other ethnic-cultural 
groups because any one ethnic-cultural community does 
not exist by itself in the everyday, ecological context or 
setting, but it exists amid other groups whose presence 
does influence social regularities of interest to community 
researchers. For instance, in studying interethnic relation- 
ships between Asian Americans and Whites, one should 
not ignore the impact of African Americans and other 
groups in the relationships. Kitano (1985) has argued that 
to understand the status of Asian Americans, the stratifi- 
cation between African Americans and Whites is also 
important to study because he conceives of Asian Ameri- 
cans as a "middleman" minority — a buffer between 
dominant and subordinate groups. 
Because of the practical problems involved in ethnic- 
cultural group research (e.g., difficulties in finding ad- 
equate samples), it may not be possible to appropriately 
study various levels of cultural complexity for ethnic- 
cultural groups. For example, research on urban Ameri- 
can Indians can be directed to the group as an aggregate, 
to members of certain tribes, or to people who live in 
certain areas of a city. As the group is further divided on 


the basis of subcultural units of cultural complexity, it 
may be increasingly difficult to find adequate sample 
sizes. This is one reason why simple rather than more 
complex levels of cultural complexity have been studied 
most frequently; however, in such research contexts, 
qualitative methods such as participant observation and 
focus group approaches can be used. The model can 
provide a practical means of categorizing research contri- 
butions and of demonstrating the state of knowledge. 
What levels of cultural complexity have guided our knowl- 
edge of ethnic-cultural groups? What research questions 
and methods have primarily been used to address which 
questions? Where are the most important gaps in the cube 
model for a particular group? Thus, the model provides a 
"context'' for understanding the state of the research for 
conducting ethnic-cultural community research. 
• Because the proposed model implies ecological flexibility 
or fluidity of one's ethnic identification and an ethnic- 
cultural community to which he or she belongs, future 
research on ethnic identity and acculturation may take 
new directions. For example, although past research has 
focused on the development of ethnic identification or 
acculturation scales for different ethnic-cultural individu- 
als (Burnam, Telles, Hough, & Escobar, 1987; Suinn, 
Richard-Figueroa, Lew, & Vigil, 1987), there is now a clear 
need for including specific contexts in which the level of 
acculturation is measured (e.g., school, home, public 
versus private places, work settings) because the effect of 
acculturation may differ in various contexts (Ethier & 
Deaux, 1990). 
Another consideration for research on ethnic identification and 
acculturation is the developmental or historical implications of 
the concepts. Most of the identity or acculturation measures have 
been developed to measure the concepts at one particular point in 
time. However, the idea of cultural complexity in the cube model 
clearly suggests that ethnic identification and acculturation both 
at the individual and the community levels can be conceptual- 
ized as having stability, duration, and permanence (Deaux, 1993). 


The static concepts of ethnic identification and acculturation must 
be reevaluated or replaced to accommodate changes due to pas- 
sage of time and significant historical events (e.g., the 1992 Los 
Angeles riots). 

A third factor in reconceptualizing ethnic identification and 
acculturation is a recognition that the behavioral focus of the mea- 
sures must be juxtaposed with other dimensions or domains, such 
as affective or cognitive aspects of ethnic identity and accultura- 
tion. For instance, it is possible that an individual "behaviorally'' 
uses English almost all the time (which most acculturation mea- 
sures indicate as the attainment of high acculturation); however, 
it could be a reflection that he or she simply spends more time in 
the English-speaking contexts and, rather, wishes to use his or 
her native language more often. Thus, the individuars affective 
level of language use needs to be assessed in order to obtain a 
more comprehensive picture of an individuars acculturation. 

Finally, the cube model can serve as a convenient way to iden- 
tify limitations in a study. Researchers who apply research find- 
ings derived from the acultural to the subcultural level are ignoring 
individual differences, whereas those who conduct research at the 
subcultural level and draw implications to the acultural level may 
be overgeneralizing. In essence, the cube model serves a heuristic 
and conceptual purpose in helping to define what kinds of re- 
search have been conducted, the appropriateness of conclusions, 
and gaps in our knowledge. 

Limitations of ttie Modei 

The proposed conceptual-methodological model has limitations. 
First, it is not intended to be the only appropriate model. Indeed, 
many other variables and dimensions can be identified as being 
important. We have presented this model only to illustrate how 
ethnic-cultural community research in the past has typically been 
based on inappropriate meta-methodological assumptions, includ- 
ing ambiguous definitions of an ethnic-cultural community. Even 
though such research has been valuable, ethnic-cultural research 
has evolved to the point where issues of heterogeneity and differ- 
ent ethnic "senses'' of community should now be addressed. 


Moreover, the model cannot intrinsically offer insights into 
the best research questions to ask, methods to use, or cultural com- 
plexities to examine. Descriptive or explanatory research each has 
certain merits; qualitative research is not necessarily less sophis- 
ticated than, or an earlier stage of, quantitative research — and 
acultural research can be as valuable as subcultural research. 
Human judgment and experience is needed to render such in- 
sights. The model offers a means of organizing and conceptualiz- 
ing ethnic-cultural research endeavors so that the types of research, 
research problems, and the implications from findings are more 
explicitly conceptualized and approached. 


Alba, R.D. (1990). Ethnic identity: The transformation of America. New Haven, CT: 
Yale University Press. 

Barzun, J., & Graff, H.F. (1985). The Modern Researcher (4th ed.). New York: 
Harcourt Brace Jovanovich. 

Bengtson, V.L., Grisby, E., Corry E.M., & Hruby M. (1977). Relating academic 
research to community concerns: A case study in collaborative effort. Journal 
of Social Issues, 33(4), 75-92. 

Berger, R, & Luckman, T. (1966). The social construction of reality: A treatise in the 
sociology of knowledge. New York: Anchor Books. 

Berry J.W., Poortinga, Y.H., Segall, M.H., & Dasen, RR. (1992). Cross-cultural 
psychology: Research and applications. New York: Cambridge University Press. 

Betancourt, H., & Lopez, S.R. (1993). The study of culture, ethnicity, and race in 
American psychology. American Psychologist, 48{6), 629-637. 

Botvin, G.J. (1986). Substance abuse prevention research: Recent developments 
and future directions. Journal of School Health, 56, 369-386. 

Brewer, M.B. (1991). The social self: On being the same and different at the same 
time. Personality & Social Psychology Bulletin, 17, 475^82. 

Brislin, R.W., Lonner, W.J., & Thorndike, R.M. (1973). Cross-cultural research 
methods. New York: John Wiley. 

Bronfenbrenner, U. (1979). The ecology of human development. Cambridge, MA: 
Harvard University Press. 

Burnam, M.A., Telles, C.A., Hough, R.L., & Escobar, J.L. (1987). Measurement of 
acculturation in a community population of Mexican Americans. Hispanic 
Journal of Behavioral Sciences, 9(2), 105-130. 


Cervantes, R.C., & Acosta, EX. (1992). Psychological testing for Hispanic 
Americans. Applied & Preventive Psychology, 1, 209-219. 

Chavis, D.M., & Newbrough, J.R. (1986). The meaning of "community" in 
community psychology. Journal of Community Psychology, 14{A), 335-340. 

Chavis, D.M., Stucky, P.E., & Wanderman, A. (1983). Returning research to the 
community: A relationship between scientist and citizen. American 
Psychologist, 38, 424^34. 

Cheung, Y.W. (1989). Making sense of ethnicity and drug use: A review and 
suggestions for future research. Social Pharmacology, 3(1/2), 55-82. 

Cook, T.D., & Campbell, D.T. (1979). Quasi-experimentation: Design & analysis 
issues for field settings. Chicago: Rand McNally College Publishing Company. 

Deaux, K. (1993). Reconstructing social identity. Personality & Social Psychology 
Bulletin, 19{1),A-12. 

Durkheim, E. (1898). Representations individualist et representations collectives. 
Revue de Metaphysique, 6, 274-302. 

Ethier, K.A., & Deaux, K. (1990). Hispanics in ivy: Assessing identity and 
perceived threat. Sex Roles, 22, 427^40. 

Evans, B.J., & Whitfield, J.R. (Eds.). (1988). Black males in the United States: An 
annotated bibliography from 1967-1987. Washington, DC: American 
Psychological Association. 

Farrell, A.D., Danish, S.J., & Howard, C.W. (1992). Risk factors for drug use in 
urban adolescents: Identification and cross-validation. American Journal of 
Community Psychology, 20(3), 263-286. 

Felner, R.D., Phillips, R.S., DuBois, D., & Lease, A.M. (1991). Ecological 
interventions and the process of change for prevention: Wedding theory and 
research to implementation in real world settings. American Journal of 
Community Psychology, 19{3), 379-387. 

Felton, B., & Shtnn, M. (1992). Social integration and social support: Moving 
beyond the individual level. Journal of Community Psychology, 20(2), 103-115. 

Golding, J.M., & Baezconde-Garbanati, L.A. (1990). Ethnicity, culture, and social 
resources. American Journal of Community Psychology, 18{3), 465-486. 

Goldman, M.S., Brown, S.A., & Christiansen, B.A. (1987). Expectancy theory: 
Thinking about drinking. In H.T. Blare & K.E. Leonard (Eds.), Psychological 
theories of drinking and alcoholism (pp. 181-226). New York: Guilford Press. 

Green, C.W, Adams, A.M., & Turner, C.W. (1988). Development and validation 
of the school interracial climate scale. American Journal of Community Psychology, 
26(2), 241-259. 

Hall, C.C.I., Evans, B.J., & Selice, S. (Eds.) (1989). Black females in the United States: 
A Bibliography from 1967-1987. Washington, DC: American Psychological 


Johnson, R.C., & Nagoshi, C.T. (1989). Asians, Asian-Americans, and alcohol. 
Journal of Psychoactive Drugs, 22, 45-52. 

Kingry-Westergaard, C, & Kelly, J.G. (1990). A contextualist epistemology for 
ecological research. In P. Tolan, C. Keys, F. Chertok, & L. Jason (Eds.), 
Researching community psychology: Issues of theory and methods (pp. 23-31). 
Washington, DC: American Psychological Association. 

Kitano, H.H. (1985). Race relations. Englewood Cliffs, NJ: Prentice Hall. 

Koegel, P., Burnam, A., & Farr, R.K. (1986). The prevalence of specific psychiatric 
disorders among homeless individuals in the inner city of Los Angeles. 
Archives of General Psychiatry, 45, 1085-1092. 

Lambert, W.E., & Taylor, D.M. (1990). Coping with cultural and racial diversity in 
urban America. New York: Praeger. 

Landrine, H., & Klonoff, E.A. (1992). Culture and health-related schemas: A 
review and proposal for interdisciplinary integration. Health Psychology, 21(4), 

Leong, FT.L., & Whitfield, J.R. (Eds.) (1992). Asians in the United States: Abstracts 
of the psychological and behavioral literature, 1967-1991. Washington, DC: 
American Psychological Association. 

Liu, W.T. (1980). Community research and the Asian American Mental Health 
Research Center. In A.K. Murata & J. Salvador-Burris (Eds.), Issues in 
community research: Asian American perspectives (Occasional Paper No. 5). 
Chicago: Pacific/Asian Mental Health Research Center. 

Loo, C, Fong, K.T., & Iwamasa, G. (1988). Ethnicity and cultural diversity: An 
analysis of work published in community psychology journals, 1965-1985. 
Journal of Community Psychology, 16, 332-349. 

Los Angeles Times. (1992). Understanding the riot. Los Angeles: Times Mirror Inc. 

Marin, C, & Marin, B.V. (1991). Research with Hispanic populations. Newbury 
Park, CA: Sage Publications. 

Maton, K.I. (1989). Towards an ecological understanding of mutual-help groups: 
The social ecology of "fit." American Journal of Community Psychology, 27(6), 

McMillan, D.W, & Chavis, D.M. (1986). Sense of community: A definition and 
theory. Journal of Community Psychology, 14, 6-23. 

Milburn, N.G., Gary, L.E., Booth, J.A., & Brown, D.R. (1991). Conducting 
epidemiologic research in a minority community: Methodological 
considerations. Journal of Community Psychology, 29(1), 3-12. 

Miles, M.B., & Huberman, A.M. (1984). Qualitative data analysis: A sourcebook of 
new methods. Beverly Hills, CA: Sage Publications. 

Moscovici, S. (1984). The phenomenon of social representations. In R.M. Farr & 
S. Moscovici (Eds.), Social representations. Cambridge, England: Cambridge 
University Press. 


Newb rough, J.R. (1992). Community psychology in the postmodern world. 

Journal of Community Psychology, 20, 10-25. 

Office for Substance Abuse Prevention (1991). The future by design: A community 
framework for preventing alcohol and other drug problems through a systems approach. 
Washington, DC: U.S. Government Printing Office. 

Olmedo, E.L., & Walker, V.R. (Eds.) (1990). Hispanics in the United States: Abstracts 
of the psychological and behavioral literature, 1980-1989. Washington, DC: 
American Psychological Association. 

Phinney, J. (1991). Ethnic identity in adolescents and adults: A review of research. 

Psychological Bulletin, 108, 499-514. 

Rappaport, J. (1977). Community psychology: Values, research, and action. New^York: 
Holt, Rinehart, & Winston. 

Rappaport, J. (1981). In praise of paradox: A social policy of empowerment over 
prevention. American Journal of Community Psychology, 9, 1-25. 

Rappaport, J. (1987). Terms of empowerment/exemplars of prevention: Toward 
a theory of community psychology. American Journal of Community Psychology, 
15, 121-148. 

Rogler, M.H. (1989). The meaning of culturally sensitive research in mental health. 
American Journal of Psychiatry, 146, 296-303. 

Root, M.RR (1992). Within, between, and beyond race. In M.RP Root (Ed.), 
Racially mixed people in America (pp. 3-11). Newbury Park, CA: Sage 

Sasao, T (1992a). Correlates of substance use and problem behaviors in multiethnic 
high school settings. Unpublished manuscript. 

Sasao, T. (1992b). Substance abuse in Asian /Pacific Islanders. In R.C. Cervantes 

(Ed.), Substance abuse and gang violence (pp. 3-23). Newbury Park, CA: Sage 

Sasao, T, & Sue, S. (1993). Toward a culturally-anchored ecological framework 
of research in ethnic-cultural commimities. American Journal of Community 
Psychology, 21,705-727. 

Schinke, S.P., Botvin, G.J., & Orlandi, M.A. (1992). Substance abuse in children and 
adolescents: Evaluation and intervention. Newbury Park, CA: Sage Publications. 

Seidman, E. (1987). Toward a framework for primary prevention research. In 
J.A. Steinberg & M.M. Silverman (Eds.), Preventing mental disorders: A research 
perspective (pp. 2-19). Washington, DC: U.S. Government Printing Office. 

Seidman, E. (1988). Back to the future, community psychology: Unfolding a theory 
of social intervention. American Journal of Community Psychology, 26(1), 3-24. 

Seidman, E. (1990). Pursuing the meaning and utility of social regularities for 
community psychology. In P. Tolan, C. Keys, F. Chertok, & L. Jason (Eds.), 
Researching community psychology: Issues of theory and methods (pp. 91-100). 
Washington, DC: American Psychological Association. 


Shinn, M. (1990). Mixing and matching: Levels of conceptualization, 
measurement, and statistical analysis in community research. In P. Tolan, C. 
Keys, F. Chertok, & L. Jason (Eds.), Researching community psychology: Issues of 
theory and methods (pp. 111-126). Washington, DC: American Psychological 

Shweder, R.A., & Sullivan, M.A. (1993). Cultural psychology: Who needs it? 
Annual Review of Psychology, 44, 497-523. 

Snowden, L. (1987). The peculiar successes of community psychology: Service 
delivery to ethnic minorities and the poor. American Journal of Community 
Psychology, 15(5), 575-586. 

Speer, R, Dey A., Griggs, R, Gibson, C, Lubin, B., & Hughey J. (1992). In search 
of community: An analysis of community psychology research from 
1984-1988. American Journal of Community Psychology, 20(2), 195-209. 

Sue, S. (1991). Ethnicity and culture in psychological research and practice. In 
J.D. Goodchilds (Ed.), Psychological perspectives on human diversity in America 
(pp. 7^6). Washington, DC: American Psychological Association. 

Sue, S., Ito, J., & Bradshaw, C. (1982). Ethnic minority research: Trends and 
directions. In E.E. Jones & S.J. Korchin (Eds.), Minority mental health (pp. 37-58). 
New York: Praeger. 

Sue, S., & Morishima, J. (1982). The mental health of Asian Americans. San Francisco: 

Suinn, R.M., Richard-Figueroa, K., Lew, S., & Vigil, R (1987). The Suinn-Lew 
Asian Self-Identity Acculturation Scale: An initial report. Educational and 
Psychological Measurement, 47, 401^07. 

Szapocznik, J., & Kurtines, WM. (1993). Family psychology and cultural diversity: 
Opportunities for theory, research, and application. American Psychologist, 
48{A), 400-407. 

Tajfel, H., & Turner, J.C. (1986). The social identity theory of intergroup behavior. 
In S. Worchel & WG. Austin (Eds.), Psychology of intergroup relations (2nd ed.). 
Chicago: Nelson-Hall. 

Tolan, P., Keys, C, Chertok, R, & Jason, L. (Eds.) (1990). Researching community 
psychology: Issues of theory and methods. Washington, DC: American 
Psychological Association. 

Triandis, H. (1992). Cross-cultural research in social psychology. In D. Granberg 
& G. Sarup (Eds.), Social judgments and intergroup relations: Essays in honor of 
Muzafer Sherif {pp. 299-243). New York: Springer- Verlag. 

Triandis, H., & Lambert, W.W (Eds.) (1980). Handbook of cross-cultural psychology: 
Vol. 1. Perspectives. Boston: AUyn & Bacon. 

Trickett, E.J. (1990). Partial paradigms and professional identity: Observations 
on the state of community psychology research. In P. Tolan, C. Key, F. Chertok, 
& L. Jason (Eds.), Researching community psychology: Issues of theory and methods 
(pp. 209-214). Washington, DC: American Psychological Association. 


Trickett, E.J., Watts, R., & Birman, D. (1992). Human diversity and community 
psychology: Still hazy after all these years. Unpublished manuscript. 

Trimble, J. (1988). Putting the etic to work: Applying social psychological 
principles in cross-cultural settings. In M.H. Bond (Ed.), The cross-cultural 
challenge to social psychology (pp. 109-121). Newbury Park, CA: Sage 

Trimble, J.E. (1990-1991). Ethnic specification, validation prospects, and the hiture 
of drug use research. International Journal of the Addictions, 25(2A), 149-170. 

Trimble, J.E., Bolek, C.S., & Niemcryk, S.J. (Eds.). (1992). Ethnic and multicultural 
drug abuse: Perspectives on current research. New York: Haworth Press. 

Turner, J.C, Hogg, M.A., Oakes, PJ., Reicher, S.D., & Wetherell, M.S. (1987). 
Rediscovering the social group: A self-categorization theory. Oxford, England: Basil 

Vega, W.A. (1992). Theoretical and pragmatic implications of cultural diversity 
for community research. American Journal of Community Psychology, 20(3), 

Watters, J.K., Downing, K., Case, P., Lorvick, J., Cheng, & Fergusson, B. (1990). 
AIDS prevention for intravenous drug users in the community: Street-based 
education and risk behavior. American Journal of Community Psychology, 1S(4), 

Watts, R.J. (1992). Elements of a psychology of human diversity. Journal of 
Community Psychology, 20, 116-131. 

Zane, N., & Sasao, T (1992). Research on drug abuse among Asian Pacific 
Americans. In J.E. Trimble, C.S. Bolek, & S.J. Niemcryk (Eds.), Ethnic and 
multicultural drug abuse: Perspectives on current research. New York: Haworth. 

Zane, N., & Sue, S. (1986). Reappraisal of ethnic minority issues: Research 
alternatives. In E. Seidman & J. Rappaport (Eds.), Redefining social problems 
(pp. 289-304). New York: Plenum. 

Author's Notes 

The preparation of the manuscript was supported in part by a 
grant from the Asian American Health Forum Inc. via a coopera- 
tive agreement with the National Center for Health Statistics (U50/ 
CCU907249) and in part by a grant from the National histitute of 
Mental Health (ROl MH44331). Correspondence concerning this 
chapter should be sent to Toshiaki Sasao, Ph.D., Department of 
Psychology, University of California, Los Angeles, 405 Hilgard 
Avenue, Los Angeles, CA 90024-1563. 



Culturally Specific 
ApproQclnes to Knowing, 
Thinking, Perceiving, and 


Kenneth H. Cushnec Ph.D. 

The term world view generally refers to the manner in which a 
cultural group perceives people and events. Understanding an 
individual's or group's world view enables one to gain a perspec- 
tive on another 's experience, on one's willingness to be open to 
other points of view, on one's ability to communicate in a manner 
that is understood by another, and on one's willingness to accept 
guidance and direction from others. Ultimately, such understand- 
ing facilitates agreeing on the meaning of certain events, includ- 
ing acknowledging that a problem exists and that a way should 
be found to approach the healing process. Gaining such an un- 
derstanding is a complex and rather extensive process about which 
volumes have been written. This chapter surveys this broad area 
in an attempt to equip professionals with broadened skills in analy- 
sis, interpretation, and communication. Deriving a world view 
results from the interaction of a conceptual system and certain 
philosophical assumptions about the world. Although these un- 
derlying assumptions are often hidden from one's conscious 
awareness, they have a significant impact on the development 
and emergence of the self and one's belief system. Concepts of 
the self are inextricably bound together and interdependent — that 


is, just as conception of the self is a cultural product, culture is a 
product of the self. Culture teaches how the self is to be construed: 
whether an individual is to be independent and central, for in- 
stance, or intimately linked with others who are looked to for 
guidance and direction. At the same time, any given culture's prod- 
ucts, practices, and perceptions are the result of activity and ef- 
fort of the individuals in that culture. 

The way that things, including the self, are conceptualized 
differs from culture to culture and from individual to individual. 
The fact that people conceptualize, however, is universal. Con- 
sideration of alternative perspectives, generally done infrequently 
by people socialized in the European or Western manner, is criti- 
cal to the understanding and delivery of health care in an increas- 
ingly pluralistic society. This paper will explore the socialization 
process in a variety of U.S. ethnic groups. 

Defining Cuiture and Related 

When people consider the concept of culture, they generally in- 
clude aspects of the world around them that have been modified 
by humans, including the physical and immaterial world. Maehr 
(1974) suggests that culture represents a group's preferred way of 
perceiving, judging, and organizing the ideas, situations, and 
events they encounter in their daily lives. Culture might involve 
the use of a particular communication system, adherence to a cer- 
tain religious belief, or the manner in which people interact in a 
society that does not value their ethnicity. Culture also determines 
the manner that people use to select information that they deem 
important or the way in which they learn how to learn the spe- 
cific content and processes determined to be essential by the group. 
Triandis (1972) differentiates aspects of culture into two cat- 
egories: objective versus subjective culture, often referred to as sur- 
face versus deep culture. That is, objective elements of a culture are 
those aspects that are tangible, visible — they can easily be exam- 
ined, felt, tasted, and so forth. The artifacts people make, the food 
they eat, and the clothes they wear are all examples of objective 


elements of a culture. Subjective elements, on the other hand, in- 
clude the less tangible, even invisible aspects of people's think- 
ing. Subjective aspects of culture may include such things as the 
values people express, the norms of behavior, their world view, 
and their attitudes — generally the things people carry around in 
their mind that are more difficult to pick up, visualize, and ana- 
lyze. Like the iceberg of which only 10 percent is visible above the 
surface of the water, there are some aspects of culture that are 
visible upon surface examination, the objective culture, so to speak. 
But like the iceberg, whose 90 percent remains hidden from view, 
there are subjective elements of culture that are also often hidden. 
These aspects of deep or subjective culture, like the hidden as- 
pects of the iceberg, are more potent and more substantial in that 
they are critical in providing support and structure for what is 
seen, as well as more dangerous in people's interactions and un- 
derstanding. It is the deep aspects of culture, or world view, that 
must be examined and understood if interactions across cultural 
boundaries are to be effective. 

Draguns (1990) points out two interrelated uses of the term 
culture. On the one hand, culture refers to the characteristics of 
people across the planet; on the other, culture refers to the social 
variations among the numerous components of a contemporary 
pluralistic society, like the United States or Australia. Significant 
cultural diversity, which can be observed close to home in almost 
any community, refers to this use of the term. Most readers of this 
chapter are primarily concerned with this latter sense of culture 
as it is represented in the composition of their clientele. 

When analyzing cross-cultural differences, a few fundamen- 
tal concepts that help to explain how specific groups of people 
come to view themselves and the world around them are impor- 
tant to consider. To begin with, it is useful to differentiate between 
emic and etic aspects of culture. "Etic" derives its meaning from 
the linguistic termphonetics, a field of study that attempts to docu- 
ment and analyze all sounds that are present in all languages in 
one meaningful framework. When used in terms of cross-cultural 
comparisons, etics refer to principles that are present in all cul- 
tures, such as communication processes, learning style differences. 


or expression of values. Cultural emics, on the other hand, refer 
to those aspects of a culture that express meaning within any one 
culture, with attention given to what people themselves value as 
important and what is familiar to them. 'Tmic'' is derived from 
the linguistic termphonemics, a field of study that documents sounds 
in a specific language that have meanings in and of themselves. 

A useful example of the emic-etic distinction may be made by 
comparing the concept "waves on the ocean or sea'' from the per- 
spective of a European American with that of a Truk Islander. 
Members of both cultures have similar conceptions of at least some 
connotations of a wave. The proposed etics here might be that 
both cultures understand the use of waves as vehicles for surfing 
and as movement reflecting the transfer of energy through the 
sea. However, in addition to this etic core, certain differences, or 
emics, exist. For European Americans, the waves may be impor- 
tant as sources of beauty and relaxation. On the other hand, the 
Truk Islander has learned to use the motion of the waves in a 
manner similar to the way a European American uses a road map. 
Hence, while one individual may perceive himself or herself to 
be lost at sea, another may be quite confident and able to use the 
patterns of the waves as street markers, so to speak. 

Tulsi Saral (1976) sums up the distinction between emics and 
etics in the following words: 

It is thus apparent that there is not absolute reality, nor is there 
a universally valid way of perceiving, cognizing, and/or thinking. 
Each worldview has different underlying assumptions. Our nor- 
mal state of consciousness is not something natural or given, 
nor is it universal across cultures. It is simply a specialized tool, a 
complex structure for coping with the environment. 

In order to adequately understand the conditions that bring a 
person to a perceived imbalance, one must understand the spe- 
cific cultural and social forces that are influential, as well as un- 
derstand how one approaches corrective action. 


Perception, Cognition, 
and Related Concepts 

Individuals who hold differing assumptions and expectations 
about the world may draw contrary inferences from the same set 
of observed phenomena or experiential events (Myers, 1988), Such 
a possibility can be explained in terms of the processes of percep- 
tion and cognition. Perception refers to what people immediately 
register on their senses — what they report seeing, hearing, tast- 
ing, feeling, and smelling. It is what people do with the stimuli 
they perceive that is of interest here. It is essential that systems 
evolve to handle the vast array of stimuli each person confronts 
on a daily basis. It is through the learned process of categoriza- 
tion — grouping similar stimuli together based on some common 
criteria — that meaning begins to evolve within the individual. It 
is by being socialized in a specific group or culture that meaning 
is given to the categories one forms in the mind. The concept dog, 
which all people perceive in the same manner, can be categorized 
as a beloved family member in the United States, as an animal to 
be avoided at all costs among traditional Muslims, or as a quite 
suitable meal by some Pacific Islanders. What must be kept in 
mind, however, is that most individuals living in a highly plural- 
istic society rarely reflect or act from one perspective only. Indi- 
viduals may be differentially influenced by their religion, gender, 
age, ethnicity, national culture, or socioeconomic status at certain 
times and in certain circumstances. Culture is complex and dy- 
namic, and at times one dimension may be more salient than 

Psychologists seem to be in less agreement about a definition 
of cognition. At best, cognition refers to the act or process of know- 
ing, perceiving, and conceiving. We can begin to explore it in terms 
of some rather loose phrases, such as "cognitive activity'' or ''cog- 
nitive functioning." It is also evident that what we are speaking 
of is inextricably bound to one's affective or emotional system, 
thereby influencing one's subjective understanding of the world. 
But focusing on cognition and the manner in which it interacts 
with the idea of culture is of most importance here. Central to any 


discussion of cognition is the issue of knowledge. And even if 
agreement could be reached about what constitutes cognition, 
what is determined to be valued and agreed-upon knowledge in 
one culture may be perceived as groundless superstition in an- 
other. Hence, we find that people become rather biased and eth- 
nocentric in their perception of the world if they hold so firmly to 
their world view that they perceive other views as less than ad- 
equate or holding inferior knowledge or practice. ^ 

Socialization and tine 
Derivation of Meaning 

How people derive meaning from the world to which they are 
born is the subject of much debate (Trevarthen, 1988). Differing 
patterns of childrearing, family lifestyle, contrasting value sys- 
tems, sociolinguistic variations, and different social orders con- 
tribute to the manner in which children are socialized. These, as 
well as other dimensions, interact to determine the manner in 
which people perceive and come to know their world, and thus 
the meaning derived from their early experiences. Behavioral 
learning theorists, for instance, suggest that certain biological 
needs that regulate the survival of the organism are innate to ev- 
ery human being and thus drive development. Social learning 
theorists stress the importance of modeling and imitation, paying 
scant attention to the origin of this capacity. Piagetian cognitive 
theory suggests that knowledge is the outcome of the interaction 
between the active individual organism and the environment. The 
child, in this sense, gains symbolic, moral, and cooperative abili- 
ties as he or she gains concepts or schemata to represent objects 
and processes in the real world. The innate intersubjectivity theory 
(Trevarthen, 1988) of sociocultural development attempts to ex- 
plain the claim that infants possess an inherent readiness to link 
their subjective evaluations of experience with those of other per- 
sons. Such a theory suggests that children start cognitive learning 
in a cooperative and imitative relationship with other more expe- 
rienced companions, contributing more to the socialization pro- 
cess than previously considered. 


Studies from Lagos, Nigeria; Edinburgh, Scotland; Madurai, 
India; and Fiji that look at the behavior of infants at birth and 
their orientation toward others confirm that although differences 
exist in the outward manifestation, certain similarities are appar- 
ent across cultures. Children at 2 years of age, for instance, show 
increasing awareness of parental standards of good and bad, as 
well as use of objects and actions. Children at this age use objects 
as instruments to do tasks for which the objects were made, per- 
form formal gestures and attitudes in social interactions with oth- 
ers, and assume social roles with pride and confidence, all in 
culturally specific ways (Trevarthen, 1988). 

Later, differing cognitive patterns emerge that determine the 
manner in which an individual derives information from one's 
environment. Students of psychology have long studied the rela- 
tionship between a given stimulus and an organism's response to 
the stimulus. Two general approaches have emerged: the stimu- 
lus-response (S-R) and the stimulus-organism-response (S-O-R) 
models. The S-R model suggests that behavior can be predicted 
by studying the functional relationships between stimuli and re- 
sponses. The S-O-R model suggests that relationships between 
stimuli and responses are best predicted from information about 
the mediating processes that occur in an organism. Cognitive style 
(also commonly referred to as learning style), proposed as one 
such mediating process, is seen as individual variations in how 
people perceive, think, solve problems, learn, and absorb and re- 
tain information and /or skills (Cushner, 1990; Kolb, 1976; Witkin, 
1962). Differing cultural experiences may result in distinctly dif- 
ferent cognitive styles. In addition, there is no consensus on what 
all cultures consider valid knowledge. We cannot understand 
knowledge without considering the culture from which it is de- 
rived. Problems then arise when individuals attempt to explain 
cognitive functioning and cognitive outcomes in terms of their 
own culturally determined frame of reference. 


Western Versus Non-Western 
Conceptions of Self 

Certain to be problematic in exploring the non- Western experi- 
ence is the tendency to use Western theoretical constructs, or to 
describe non- Western phenomena in comparison to a Western 
norm. The potential exists to miss the reality that most of the 
world's peoples do not adhere to or accept the same underlying 
assumptions. In addition, although not entirely within the scope 
of this chapter, problems concerning measurement issues when 
comparing and contrasting differing world views cannot be ig- 
nored. These problems have proved to be most vexing to the ac- 
tivities of cross-cultural researchers and are those that allow 
ethnocentric bias to easily reemerge. Researchers holding to a par- 
ticular world view may show a substantial sensitivity to other 
cultures at the philosophical level, but when it comes to the ac- 
tual work of cross-cultural research, they may resort to "research 
business as usual.'' This sometimes results from an incomplete 
understanding of cross-cultural issues but more often is probably 
related to the general lack of research tools to do the work re- 
quired. In the upshot, the ''normal" instruments are resorted to, 
with the result that all non-Western cultures are compared to 
Western culture, the implicit standard. 

Differing cultural patterns of knowing also create divergent 
forms of selfhood as the individual strives to achieve a dynamic 
balance between psychic demands and social-cultural require- 
ments within a particular context. Long ago, a comprehensive 
typology of Western versus non- Western social systems was de- 
veloped by Kluckhohn and Strodbeck (1961). They organized the 
variety of social systems around five questions: (1) What is the 
character of innate human nature? (2) What is the relationship of 
persons to nature and supernature? (3) What is the temporal fo- 
cus of human life? (4) What is the modality of human activity? (5) 
What is the modality of a person's relationship to other persons? 

Pedersen (1979) identified several critical issues that separate 
Western from non- Western psychology. In the Western perspec- 
tive, for instance, accepted knowledge is acquired primarily 


through the intellectual process. This, as Berman (1989) suggests 
in his criticism of the study of history, is often a set of abstractions 
and a clutch of formulas, devoid of personal meaning and rel- 
evance. From a non- Western perspective, knowledge may be at- 
tained through many additional forms, including identity through 
close association between two individuals (e.g., teacher /mentor 
and student), from ancestral contact, or, as Berman (1989) again 
suggests, through intuition or a ''body literacy'' that operates more 
or less at an unconscious level. In some contexts, knowledge 
gained through intuition is more highly valued. One's identity 
from a Western perspective is considered highly individualistic, 
as well as private. The healthy individual is seen as separate from 
and unique among others — so much so that, in some cases, if one 
is seen as too much like others, he or she may be viewed as lack- 
ing in personality. This can be observed by the fact that many 
Western (i.e., Indo-European) languages have only one pronoun 
for the first and second person singular (I, thou; ich, du). In most 
non- Western languages there is more than one word for I or thou. 
In Japanese, for instance, there are at least two words, which dif- 
fer according to status, gender, public /private use, and so forth. 
One's identity, so to speak, is attached to a web of others. 

The emphasis in Western society, too, is often on surface quali- 
ties. Typically, however, in the non- Western world attachment to 
an individual or to the self can be minor, and in some cases per- 
ceived as a handicap. Individual development may be perceived 
as a lower stage of enlightenment and a diversion from more im- 
portant, spiritualistic goals of existence. Therefore, collective ori- 
entation develops, and one strives to attain a balance of the self 
and the non-self, between internal and external realities. 

In the West, there is strong socialization toward assertive in- 
dependence: speaking out, direct communication, and unrestricted 
expression of anger and other feelings. In most non- Western col- 
lective societies, individuals are socialized toward interdepen- 
dence with others, especially extended family members. The self 
is second to others. Rather than directly confronting others, there 
is a tendency to "smooth over" problems and to strive to find a 
balance. Among native Hawaiians, for instance, the practice of 


ho'opono'pono (literally "to make right'') is a form of conflict reso- 
lution that involves extended-family members in the resolution 
of problems that emerge in a group setting. Markus and Kitayama 
(1991) propose that in many collectivist societies, mature adults 
are those who are able to think of their own personal needs being 
secondary to the needs of the group. Unlike in the West, giving in 
to another person's decision is not a sign of passivity or weakness 
gesture but of tolerance, self-control, flexibility, and maturity 

In the Western sense, development is seen as ending in adult- 
hood. In a non- Western perspective, development is seen as on- 
going, and the adult is seen as having achieved the preliminary 
groundwork to begin spiritual development. In China, for in- 
stance, a son's identity is tied to perpetual dependence on his 
parents. In Japan, there is a developmental curve that allows the 
most freedom in infancy and old age, but imposes the most con- 
trol during the middle range of years — nearly the opposite of de- 
velopmental theories in Western cultures. Western cognition is 
based on the objects of consciousness, whereas non- Western psy- 
chologies tend to emphasize consciousness itself. Whereas West- 
ern experience tends to discount phenomena not observed or 
measurable, non- Western psychology emphasizes consciousness 
itself and stresses that people are more than their physical body 
An extrasensory self allows learning to occur through uncover- 
ing information that already exists in the depths of our nature but 
that is ordinarily blurred. Such sources of knowledge may include 
contact with one's ancestral spirits. 

In the Western preoccupation with self and rational, analytic 
thought, one tends to find a sense of encapsulation and isolation 
from other individuals, from the environment, and, in many cases, 
from a God. Such an orientation of the self is significantly differ- 
ent from that of many other of the world's peoples. In direct op- 
position are the more traditional orientation and non- Western 
philosophies that stress an interconnectedness and interdepen- 
dence between and among people, community, and the environ- 
ment. Self and the social setting are inextricably bound. Such is 
the case among many of the subordinated or disenfranchised 
groups in the United States. Insight into the world views, per- 


spectives, experiences, and behaviors of major ethnic /racial 
groups in the Nation will serve to assist health professionals in 
understanding the reasons behind people's afflictions as well as 
to facilitate the healing process, both in terms of improving inter- 
personal dialog and in terms of exploring culture-specific thera- 
pies. Intranational differences may best be understood by looking 
closely at research in the international arena. The following sec- 
tion of this chapter offers a survey of international cultures fol- 
lowed by a brief summary of intranational subordinate groups. 

Culture-Specific World Views 

In any intercultural encounter, one must separate out individual 
differences in personality from group or national characteristics. 
Perhaps the premier exploration of national characteristics was 
conducted by Geert Hofstede (1980, 1983), a Dutch psychologist, 
who examined over 116,000 IBM employees in more than 50 coun- 
tries (Goodman, 1994). Surveying all levels of the organization, 
from unskilled workers to top managers, and matching charac- 
teristics such as job category, age, and gender, Hofstede identified 
four dimensions of national culture that can serve as a basis for 
comparing the dominant value systems between national cultures. 

In this analysis, it is important to note that Hofstede exam- 
ined the relationship between nationality and mean value scores. 
Focusing on such a relationship meant that the country, not the 
individual respondent, became the unit of analysis. Thus, the di- 
mensions derived from the research were ecological dimensions 
of collective national cultures, not dimensions of individual per- 
sonality It is, however, appropriate to consider the dimensions of 
national culture as examples and measures of the ''personality'' 
of the culture. Such knowledge can be useful not only for under- 
standing the dimensions along which differences may arise, but 
also for projecting where problems in communication, understand- 
ing, and approaches to therapy might be present between inter- 
national populations in a nation. 

The four dimensions of national culture Hofstede identified 
and how they might influence behavior are reviewed below. 


Power Distance 

Power distance refers to the degree to which a society accepts the 
idea that power is to be distributed unequally among members of 
their population. The more this is accepted, the higher the 
country's ranking in power distance. In such societies, age is highly 
respected, information tends to flow from teacher to student or 
from authority figure to subordinate, and little dialog occurs across 
status lines. A culture high in power distance, as Draguns (1990) 
suggests, would reflect an emphasis on the therapist or health 
practitioner in the role of expert. Compliance with the directions 
or values offered by the professional would take precedence over 
such variables as personal growth, self-discovery, and insight. 
Confrontational techniques, such as group encounters and other 
forms of group therapy, would have low priority. On the other 
hand, a low power distance culture would emphasize self- 
improvement groups while deemphasizing the differentiation be- 
tween client and therapist roles. 


Individualism refers to the degree to which a society believes that 
an individual's beliefs and actions should be independent of col- 
lective thought and action. In societies that are strong on collec- 
tivism, there is a solid sense of respect for tradition and the group. 
Individuals find more satisfaction working with a group for a 
collective goal rather than working individually for their own 
achievements. Individuals typically do not call attention to 
themselves. Individualistic cultures tend to favor insight-oriented 
therapies, with a relatively distant professional relationship be- 
tween patient and professional. In contrast, in collectivist cultures, 
a relatively close, expressive, and personal relationship between 
professionals and clients would predominate. Issues of self- 
actualization and discovery would be deemphasized, while at- 
tempts to alleviate suffering would prevail. 



Uncertainty Avoidance 

Uncertainty avoidance refers to the degree to which a society feels 
threatened by ambiguous situations and tries to avoid them by 
providing rules and refusing to tolerate deviance. The more a so- 
ciety accepts this idea, the higher its ranking in uncertainty avoid- 
ance. Societies that are high on this dimension are characterized 
by high structure, precise plans of operation, detailed assignments, 
and adherence to a schedule set up well in advance. Cultures high 
on uncertainty avoidance would stress the "scientific" explana- 
tions of cause and cure. Behavior modification approaches with 
rather precise objectives might be successful. Those high on this 
dimension would have a tendency to avoid most counseling or 
psychotherapeutic approaches, considering them to be rather 
unpredictable and inefficient. If used, successful psychotherapy 
might be viewed as a means to transmit scientific and medical 
information to patients. 


Masculinity refers to the degree to which a society focuses on the 
traditionally masculine values of assertiveness, task achievement, 
and the acquisition of things as opposed to the quality of life. 
Hofstede found that the more a nation is characterized by mascu- 
line values, the greater is the gap between the values espoused by 
men and women in that nation, and the more gender segregation 
in career choice, with males avoiding feminine occupations. In 
masculine cultures, a therapist or health worker would be viewed 
as representing society's side instead of the aspirations of the cli- 
ent. Central values in working with patients or clients high on 
this scale would revolve around responsibility, adjustment, and 
conformity. In more feminine cultures, a therapist or health worker 
would take the side of the individual against the society and would 
emphasize such values as expressiveness, empathy, and creativity 
The relevance of Hofstede's research (which was conducted 
in work-related settings) to health care and other settings is based 
on the assumption that role patterns and value systems in a soci- 
ety are carried forward from the home and family to work and 
other dimensions of life. In order to appreciate how the four 



dimensions can affect interactions, it is helpful to examine extreme 
differences between the dimensions, while recognizing that in 
many countries the situation may be closer to the center. How- 
ever, by looking at extreme cases, we can best identify situations 
that can create cross-cultural difficulties. 

The research findings show a strong correlation between 
power distance and individualism /collectivism (Hofstede, 1980). 
Societies that are high on power distance also tend to be high on 
collectivism; this is due to the fact that both low power distance 
and high individualism correlate with national wealth. When na- 
tional wealth is statistically controlled for, the correlation disap- 
pears. Countries high on individualism and low on power distance 
include the United States, Australia, Great Britain, The Nether- 
lands, Canada, and New Zealand. Denmark, Sweden, Switzer- 
land, Germany, Ireland, Norway, Finland, Israel, and Austria are 
a bit less individualistic and lower on power distance. 

Countries relatively high in power distance and high in indi- 
vidualism are Italy, Belgium, South Africa, France, and Spain. 
Costa Rica was the only country in the study that had low power 
distance and low individualism. The countries that were highest 
in power distance and collectivism included Guatemala, Panama, 
Venezuela, Ecuador, Colombia, Indonesia, Pakistan, Peru, Taiwan, 
Singapore, Korea, Thailand, the nations of West Africa, Hong 
Kong, and Mexico. 

Subgroups in the United States: 
European-American Perspective 

As early as 1955, Mason examined major studies on acculturation 
undertaken by anthropologists to determine what constituted 
"mainstream'' American culture. Defining this culture as Euro- 
pean, Caucasian, and primarily operative in the Western Hemi- 
sphere, he identified certain values, beliefs, customs, desired 
behaviors, and dispositional factors that were subsequently de- 
fined as necessary for achieving in American society. According 
to this analysis, the ''typical'' American is exposed to socializa- 
tion practices that emphasize (1) the importance of the individual 


over the group; (2) the idea that nature can and must be controlled 
and harnessed for human advantage; and (3) that the ultimate 
successful personality is one that is self-reliant, industrious, thrifty, 
highly motivated, and quite competitive. Also valued are punc- 
tuality, freedom of expression, future orientation, equal opportu- 
nity, social democracy, and forthright, direct action. 

A geo-anthropological explanation of such an orientation 
might consider the evolutionary past when humans migrated from 
the African continent to the more temperate regions of Europe 
and Asia. Leaving a relatively mild environment, v^here food 
sources were predictable and available year-round and extremes 
in weather were minimal, humans suddenly found a relatively 
hostile and unpredictable context with extremes in temperature 
and weather, unpredictable storms, and seasonal growing cycles 
in which to adapt or perish. Those who were able to collect and 
store an abundance of foodstuffs to last through a winter, as well 
as modify their environment to provide protection from extremes 
of weather, survived and passed on their genes. Such circum- 
stances may well explain the learned value orientations identi- 
fied above and the behaviors of many present-day European 
Americans. However, significantly different world views exist. 

The sections below describe relatively recent historical expe- 
rience and how it may influence perspectives and orientations of 
some of the major U.S. ethnic groups. These are presented with 
the hope that health care providers and community personnel 
might begin to explore how such perceptions influence the ex- 
pression of a problem as well as the delivery of health care. Con- 
siderable care should be taken, however, while reading the 
following sections. There may be a tendency to create stereotypes 
whenever any culture- specific information is presented about any 
group. Yet any discussion about cultural differences must include 
specific details. Readers should keep in mind that there is often 
greater variation within a given group than between specific 
groups. The information provided below is presented as a means 
to begin discussion and investigation of the particular groups with 
which individuals interact. 


African-American Perspective 

African Americans, like many other subordinated or disenfran- 
chised groups in the United States, have not been permitted full 
entry into mainstream society. From their earliest contacts in slave 
quarters to the experience in the urban centers today, African 
Americans have been forced to selectively maintain and /or de- 
velop an orientation, interactive style, and cultural practice that 
both protects them from harsh social forces and establishes a pat- 
tern for survival. Africans arrived in the Americas with an already 
extended sense of community, established in their highly struc- 
tured tribally oriented communal societies, where members de- 
pended on each another for physical protection and support, as 
well as for assistance in economic ventures. In their new environ- 
ment, a similar configuration developed — this time, however, 
predicated on skin color and experiential communality rather than 
tribal affiliation. The African- American kinship network is thus 
multigenerational and regularly consists of relatives, friends, and 
neighbors. Through such a network, African- American individu- 
als and their nuclear family system are able to give and receive 
emotional, physical, psychological, and social support. In fact, it 
is this sense of mutual responsibility that promotes the practice 
of informal adoption of dependent children or the inclusion of 
financially insecure elderly relatives of a single-parent family as a 
part of a household. 

As a part of this group orientation, or perhaps as a result of it, 
many African Americans developed a unique affective and per- 
sonal orientation that manifests itself in a particular approach to 
social cognition, interpersonal interaction, use of nonverbal com- 
munication, and reliance on external referents for information and 
advice. African Americans, therefore, may attend to the affective 
features of others more than do their European- American coun- 
terparts, and thus may derive much more meaning from reading 
a person's facial and other nonverbal cues. African- American in- 
dividuals are more likely to touch one another during conversa- 
tions than are their European- American counterparts, and they 
give more attention to body language (e.g., hand gestures, body 
stance, styles of walking). 


Social distance, involving the expanding and contracting of 
physical space surrounding an individual, is an area where dif- 
ferences in interpersonal style between African Americans and 
European Americans are most evident. Hall (1966) suggests that 
the perception of social cues, ideas, and attitudes is affected by 
the amoimt of physical separation demanded by the individual 
for social interaction. Those who permit individuals to come close 
gather one sort of information, while those who demand greater 
separation receive other types of cues. Studies of adults have noted 
closer social distance preference among African Americans (Bauer, 
1973; Hall, 1966). There may, thus, be a tendency to interpret in- 
terpersonal interrelationships and to perceive emotional climates 
differently between the two groups. 

A high social sensitivity to others among African Americans 
may also be the result of the fact that individuals are socialized 
from a very early age to be wary of people as well as systems in 
their environment (Shade, 1978). There is also a "culture" of op- 
pression or inferiority in which every African American is sub- 
merged. The idea that European views predominate and that 
others, including African perspectives, are unacceptable is linked 
to the perception that if one possesses less than desired physical 
traits one also displays intellectual inferiority. Ogbu (1979) and 
Howard (1980) suggest that this type of environment produces a 
type of psychobehavioral modality common to colonized or caste- 
dominated people around the world. This modality may be char- 
acterized by a high degree of self-hatred, overidentification with 
those in power, hostility, aggression, and generally inadequate 
development of the motivational, cognitive, and intellectual skills 
necessary for the mainstream society. A cultural byproduct, those 
authors assume, is a preoccupation with and group focus on con- 
cepts of freedom and equality. As a result, there appears to be a 
basic cultural consensus on what represents trustworthiness that 
is determined by reading nonverbal rather than verbal cues (Shade, 
1989). The idea that people within one's environment should be 
approached with caution and a sense of distrust is vital to the 
survival of a group of people who live in an urban society and in 
a society with a dislike of people based on skin color. This helps 


to prevent one from becoming a victim and might explain the 
tendency to avoid direct eye contact with others, especially those 
in positions of authority. 

From the combined African, European, and colonized experi- 
ences developed a particular world view, a perspective on life, 
forms of personal expression, and mechanisms for survival that 
have come to be recognized as a unique cultural pattern. 

Native-American Perspective 

Although Native Americans can trace their origins on the conti- 
nent for tens of thousands of years, their more recent experiences 
mimic those of African Americans in many ways. And like other 
groups, the behavior patterns practiced by many Native- Ameri- 
can children find their origin in a number of important values 
that are subtly, informally, and often unconsciously taught. Val- 
ues that seem to be important among the various Native Ameri- 
cans include such things as generosity and sharing, cooperation 
and group harmony, differing concepts of time, and a different 
orientation toward ownership and property. 

Native- American children, for instance, tend to be more self- 
directed and self-disciplined than their European American coun- 
terparts. Many Native- American children are, from an early age, 
given greater freedom to explore their world, to experiment, to 
question the old, and to try to invent the new than are their Euro- 
pean-American counterparts. They may thus develop greater self- 
directed discipline early in life. For most European Americans, 
obedience to some other authority figure, as well as self-direction, 
is taught later in life. For many Native- American children, there- 
fore, discipline does not mean something as narrowly defined as 
obedience but is connected in a powerful way to the emergence of 
self-discipline. As a consequence, traditional Native- American 
childrearing practices are often labeled as permissive compared with 
European- American practices. 

Respect for the dignity and autonomy of the individual is 
highly valued by Native Americans, and so children are taught 
not to interfere in the affairs of others. Thus, noninterference is 
regarded by the Native American as normal, often resulting in 



the perception that European Americans are prying into someone 
else's business. Hence, a conflict may result when Native Ameri- 
cans resist and resent the involvement of outsiders in their af- 
fairs. Native- Americans prefer a more observational learning style, 
in contrast to the European- American preference for verbal learn- 
ing. In most Native- American families, the child is a revered mem- 
ber of the unit and is welcomed as a spectator and participant in 
many family and community events. This constant and close prox- 
imity to the actions of many others provides the child with a sense 
of belonging and a valuable opportunity to become familiar with 
a wide range of simple, as well as complex, activities. In addition, 
the child often comes under the care and guidance of a variety of 
other individuals — parents, family friends, other caregivers, and 
older siblings — and, as such, will have multiple opportunities to 
participate in a number of tasks. European- American children, 
on the other hand, do not usually participate with such a wide 
range of individuals and thus tend to learn about things through 
verbal exchanges of listening and questioning. This act of ques- 
tioning, while expected and required among European Ameri- 
cans, is often perceived as an intrusion by many Native Americans. 
Native- American children thus learn the customs and skills 
of their society by sharing directly in the activities of others — 
verbal instructions are neither offered nor required, because ob- 
servation makes verbal direction redundant. Philips (1983) 
suggests that it is not that Native- American children are nonver- 
bal (as is oftentimes suggested) or unable to use the verbal chan- 
nel, but that different cultures allocate the use of verbal plus 
nonverbal systems differently In summary. Native Americans tend 
to be quite skilled in nonverbal communication as well as in pro- 
cessing visual and spatial information, while being less skilled 
and displaying a lower frequency in verbal coding. Native Ameri- 
cans also tend to be more holistic in their processing of informa- 
tion, have a preference for communal or small-group learning and 
interaction with a minimum of verbal interchange required for 
both satisfaction and success, and prefer a rather informal setting 
that allows for much freedom of movement (Pepper & Henry, 1989). 


Even though extensive variation exists among the Native 
American peoples (the Bureau of Indian Affairs recognizes 478 
different tribes), some commonality in values can be identified. 
Bryde (1972) developed a list of basic value differences between 
Native Americans and European Americans by asking Native 
Americans how they are different from those in the dominant 
culture. This list is quite informative and helps to explain a par- 
ticular world view: 

• Native Americans have a tendency to be present rather 
than future oriented. 

• Native Americans have a relative lack of time conscious- 
ness. In fact, many tribal languages have no word for 
time, believing things that are meant to be will ultimately 
be accomplished, which is in stark contrast to the Euro- 
pean-American notion of time (note the number of words, 
phrases, and proverbs that denote reference to time in 
American English). 

• Native Americans have a respect for age rather than a 
youth orientation, a cooperative versus a competitive 
orientation, and a desire for harmony with nature instead 
of its conquest. 

• Native Americans emphasize generosity and sharing 
versus personal acquisition of material goods. 

Trimble (1981) relates an example of how the Indian receives 
in order to give away. He recalls a Lower Elwha Indian woman in 
Washington's Olympic Peninsula region who had sent her grand- 
son to school with a big lunch. The non-Indian teacher, assuming 
that the lunch had been prepared for the student, made the child 
eat the entire lunch. What the teacher failed to realize was that, 
contrary to her expectations, the child had brought the lunch to 
share with others. When the child returned home, he asked the 
grandmother not to fix that kind of lunch again. 

Asian-American Perspective 

There are differences between the experiences of African Ameri- 
cans and Native Americans and those of Asian Americans in this 


country. First, Asian- American immigration to the United States 
was voluntary. Second, ties with a home country remained strong. 
For a variety of reasons, Japanese and Chinese communities re- 
mained rather close and were able to provide an alternative to 
mainstream society for newly arriving immigrants. Finally, most 
Asian groups were able to bypass a long dependence on the Fed- 
eral Government, except perhaps the more recent immigrants from 
Southeast Asia (Kitano & Matsushima, 1981). 

Nonetheless, Asian Americans share many similarities with 
other ethnic /racial groups through encounters with prejudice and 
discrimination. Most Americans do not know that Asians in the 
United States often suffered inhumane treatment. Large numbers 
of Chinese immigrants had come to the United States as early as 
1840. By 1869, when competition for jobs had increased, they were 
easily and quickly made scapegoats for the employment prob- 
lems. In 1882, the Chinese Exclusion Act was passed, forbidding 
further immigration by people from China, leading to the phrase 
''not a Chinaman's chance," which was coined during these hos- 
tile times. The Japanese began to come to the United States in 
large numbers in the 1890s. Not long after their arrival, the Alien 
Land Act of 1913 was made law, which forbade aliens to own 
land, thereby officially depriving many of the possibility of se- 
curing a place in the "American dream." Discrimination and preju- 
dice against Japanese was never greater than during World War 
II, when over 100,000 Japanese Americans were put into concen- 
tration camps. 

Differences, too, exist in the manner in which Asian Ameri- 
cans are perceived by the majority culture. While immigrants from 
European backgrounds who retain an accent are often viewed as 
sophisticated and continental, Asians with an accent tend to be- 
come the subject of ridicule and satire ("rots of ruck"). The fact 
that most Asian Americans tend to live in California or Hawaii 
and thus interact less frequently with the majority of Americans 
contributes to the persistent stereotypes and negative images. 
Many third- and fourth-generation Asian Americans are still 
treated as foreign nationals or visitors in their own country. As a 
result, a variety of responses emerged to deal with the many 


experiences faced by the Asian- American community. Some mem- 
bers became acculturated and attempted to assimilate, some re- 
acted to the discrimination and prejudice they experienced by 
retreating into their own ethnic enclaves, and still others returned 
to their homelands. Relatively few, however, turn to mainstream 
social and counseling agencies for support. 

The Asian immigrants who came to the United States brought 
with them many of the cultural characteristics and traditions that 
are still seen today. Not unlike most other immigrants who came 
to this country, in response to external threats and prejudice indi- 
viduals tended to isolate themselves from the society at large. As 
a result, an already complex culture was reinforced and retained. 
Following is an attempt to briefly summarize Japanese and Chi- 
nese cultural values (Sue, 1981). 

In the traditional family, age, sex, and generational status are 
the primary determinants of role behavior. The roles of family 
members tend to be rather rigidly defined and quite resistant to 
change. Ancestors and elders are afforded great reverence and 
respect, as is the father, the traditional head of the household, 
whose authority is generally unquestioned. Roles are well estab- 
lished and adhered to, with a son's primary duty that of being a 
good son (being a good husband and father is secondary to being 
a good son), and with females being subservient to males and 
expected to perform their domestic duties without fail. 

Conflicts within families are generally kept to a minimum due 
to the highly structured role relationships that allow for few de- 
viations. Considerable effort is expended to avoid confrontations 
with others and to approach problems as indirectly as possible. 
Restraint of possibly disruptive behaviors and strong, especially 
negative, emotions is greatly emphasized. This may help to ex- 
plain the common experiences of somatization — reporting psy- 
chological problems in terms of bodily illness (Draguns, 1990). 

The welfare and integrity of the family is of extreme impor- 
tance. Deference to this collective, above all else, reflects credit on 
the whole family. So important is the emphasis on the family's 
reputation and ''saving face'' that problems tend to be handled 
within the family as much as possible. Public admission of prob- 


lems is suppressed. Guilt and shame are the primary means used 
to reinforce obligation to family values. From a very early age, 
children who attempt to act out on their own and display their inde- 
pendence are told that they are selfish and inconsiderate of others. 
Ho (1994) identifies five common themes that emerge from a 
review of the literature on Asian cultures: (1) collectivism, (2) reci- 
procity, (3) other-directedness, (4) maintenance of harmony and 
avoidance of open conflicts, and (5) conformity, all of which sup- 
port a collective, interdependent orientation. Increasing demands 
toward assimilation and acculturation, especially in such a highly 
individualistic nation as the United States, place many Asian 
Americans in extreme culture-conflict situations. For example, 
restraint of strong emotion is highly valued in Asian culture. From 
a European- American perspective, such an individual may be 
perceived as passive and inhibited. The negative connotation of 
such values in the mainstream society results in many Asian 
Americans becoming quite confused. Thus the stage is set for a 
variety of conflicting responses. 

Hispanic-American Perspective 

While it is often useful to identify dimensions of culture that char- 
acterize a group, care must be taken to avoid the tendency to cre- 
ate stereotypes about a group of people. Differences within a group 
can be greater than the differences between groups. Such is cer- 
tainly the case with those identified as Hispanic /Latino by the 
dominant cultural group. 

The fastest growing ethnic group in the United States is com- 
posed of those who identify themselves as having Mexican heri- 
tage — roughly 8 percent of the U.S. population. But those 
identifying themselves as sharing the "Hispanic'' identity, those 
sharing a common history with people of Spanish origin and de- 
scent, include far more. Thus, "Hispanic'' denotes ethnicity, lan- 
guage, family name, or ancestry (Sue, 1981). More important, even 
though members of this "Hispanic" group are often lumped into 
one group by outsiders because they do share similar language, 
values, and customs, significant differences between groups 


(e.g., Puerto Ricans, Central Americans, Mexicans) exist, so that it 
makes more sense to conceptualize Hispanic /Latino culture as 
an aggregate of distinct subcultures, each emanating from a dif- 
ferent geographical area. 

A shared agrarian background and a strong commitment to 
Catholicism, as well as many family structures based on rather 
rigidly defined sex-role distinctions, have influenced many of the 
cultural patterns evident today For instance, Hispanic /Latino 
Americans strongly identify with their communities, families, or 
ethnic groups. Extended family and kinship networks, as well as 
strong affiliations with a companion's parents, are highly re- 
spected: success depends on cooperative efforts. In addition, the 
extended family structure may offer a stress-resistant quality, 
which may explain Hispanic Americans' underutilization of pro- 
fessional health and counseling services. This strong group ori- 
entation is especially evident in educational and related settings, 
where individuals may avoid standing out from the group. 

In the social context, Hispanic Americans tend to be more sen- 
sitive to others' feelings, to stand closer to others, to avoid eye 
contact, and to touch one another more often than European 
Americans would expect. Clearly delineated status and role defi- 
nitions predominate within the group. The perception of the male's 
being biologically superior is perhaps exaggerated by the com- 
monly used term "macho" (Spanish for "male") which is used 
among many Hispanic Americans as a flattering term to denote 
masculinity. It connotes physical strength, sexual attractiveness, 
virtue, and potency. At a more subtle level of analysis, however, 
"real" masculinity among Hispanic Americans involves dignity 
in personal conduct, respect for others, and love for the family 
and for children (Sue, 1981). 

Related to this are the relatively clearly delineated status and 
role definitions among related individuals, as well as the person- 
ality traits most often identified among Hispanic Americans 
(Shade, 1989). Such traits, including a tendency toward orderli- 
ness, less aggressive interaction, active internal control, and the 
need to have close interpersonal relationships, may all influence 
power relationships and the manner in which one approaches 
health care and interpersonal guidance. 


Cross-Cultural Interactions 
in tine Delivery of Health 
and Counseling Services 

The implications of understanding culture-specific ways in which 
people perceive their world are significant when provider and 
client are from different backgrounds. This is especially problem- 
atic in the United States, where individuals may hold bi- or 
multicultural status, with the underlying influences of assimila- 
tion and acculturation. A given bicultural client may think, act, 
and feel more like a European American than his or her other 
identity when dealing with issues related to work or school, but 
may respond more closely to his or her ethnic identity when is- 
sues are related to home and community One must consider that 
the majority of individuals identifying with a group other than 
the majority are members of both the majority and the ethnic/ 
racial culture. Second, the degree to which a specific identity is 
salient may vary with an individual's gender, age, and genera- 
tional status across context, time, and issue. Third, it seems rea- 
sonable to assume that those whose cultural identity is more with 
their ethnic group than with the mainstream will benefit from an 
approach that is more culture specific (see Pedersen, Draguns, 
Lonner, & Trimble, 1981; Sue, 1981). 

Illness behavior of any kind may be best understood and more 
readily treated in its cultural context. Ilola (1990) reviews the case 
of a Hmong refugee whose death was attributed to "Hmong sud- 
den death syndrome'' or "nightmare death," an inexplicable cause 
of death that occurs during sleep and is found most frequently 
among Hmong but also occurs among Laotians, Cambodians, and 
Vietnamese. In such cases, generally, no evidence of biological or 
pathological etiology is found. Rather, somatization of psycho- 
logical distress caused by survivor guilt, severe culture shock, and 
combat-related experiences are thought to be the cause. Even with 
bilingual translators. Western psychotherapy seems not to help. 
Instead, a well-respected shaman from the local community can 
help bring the problem under control in a relatively short time. 


For the health care provider or community worker, the best 
tool to have may be that of empathy — ^but empathy may not travel 
well across the cultural chasm (Draguns, 1973, 1990). Perhaps the 
ideal approach for those working with a diverse clientele is to 
combine personal sensitivity with cross-cultural knowledge and 
the ability to respond and adapt to the specific cultural context 
and circumstances in which they practice. 


Bauer, E. (1973). Personal space: A study of Blacks and Whites. Sociometry , 36, 

Berman, M. (1989). Coming to our senses. New York: Simon & Schuster. 

Blassingame, J. (1979). The slave community. New York: Oxford University Press. 

Bryde, J.F. (1972). Indian students and guidance. Boston: Houghton Mifflin. 

Cushner, K. (1990). Cross-cultural psychology and the formal classroom. In R. 
Brislin (Ed.), Applied cross-cultural psychology. Newbury Park, CA: Sage 

Draguns, J. (1973). Comparison of psychopathology across cultures: Issues, 
findings, directions. Journal of Cross-Cultural Psychology, 4, 9-47. 

Draguns, J.G. (1990). Applications of cross-cultural psychology in the field of 
mental health. In R. Brislin (Ed.), Applied cross-cultural psychology. Newbury 
Park, CA: Sage Publications. 

Goldstein, R. (Ed.) (1971). Black life and culture in the United States. New York: 
Thomas Crowell. 

Goodman, N. (1994). International education at the university. In R. Brislin & T. 
Yoshida (Eds.), Improving intercultural interactions: Modules for cross-cultural 
training programs. Newbury Park, CA: Sage Publications. 

Hall, E. (1966). The hidden dimension. New York: Doubleday 

Ho, D.Y.F. (1994). Toward an Asian social psychology: Relational orientation. In 
U. Kim & J. Berry (Eds.), Indigenous psychologies: Experience and research in 
cultural contexts. 

Hofstede, G. (1980). Culture's consequences: International differences in work-related 
behaviors. Beverly Hills, CA: Sage Publications. 

Hofstede, G. (1983). Dimensions of national cultures in fifty countries and three 
regions. In J. B. Deregowski, S. Dziurawiec, & R.C.Adams, {Eds. ),Expiscations 
in cross-cultural psychology (pp. 335-355). Lisse, The Netherlands: Swets and 


Howard, J.H. (1980). Toward a social psychology of colonialism. In R. Jones (Ed.), 
Black psychology. New York: Harper and Row. 

Ilola, L. (1990). Culture and health. In R. Brislin (Ed.), Applied cross-cultural 
psychology. Newbury Park, CA: Sage Publications. 

Kitano, H.H.L., & Matsushima, N. (1981). Counseling Asian Americans. In P. 
Pedersen, J. Draguns, W. Lonner, & J. Trimble (Eds.), Counseling across cultures: 
Revised and expanded edition. Honolulu, HI: East-West Center. 

Kolb, D. (1976). Learning style inventory. Boston: McBer. 

Maehr, M.L. (1974), Sociocultural origins of achievement. Monterey, CA: 

Markus, H.R., & Kitayama, S. (1991). Culture and the self: Implications for 
cognition, emotion, and motivation. Psychological Review, 98(2), 224-253. 

Mason, L. (1955). The characteristics of American culture in studies of 
acculturation. American Anthropologist, 57, 1264-1277. 

Myers, L.J. (1988) Understanding an Afrocentric world view: Introduction to an optimal 
psychology. Dubuque, I A: Kendall /Hunt. 

Nobles, W.W. (1980). African philosophy: Foundations for Black psychology. In 
R. Jones (Ed.), Black psychology. New York: Harper and Row. 

Ogbu, J. (1979). Minority education and caste. New York: Academic Press. 

Pasteur, A.B., & Toldson, I.E. (1982). Roots of soul: The psychology of Black 
expressiveness. Garden City, NY: Anchor Press /Doubleday. 

Pedersen, P. (1979) Non- Western psychology: The search for alternatives. In A. 
Marsella, R. Tharp, & T. Ciborowski (Eds.), Perspectives on cross-cultural 
psychology. New York: Academic Press. 

Pedersen, P., Draguns, J., Lonner, W, & Trimble, J. (1981). Counseling across cultures: 
Revised and expanded edition. Honolulu, HI: East-West Center. 

Pepper, EC, & Henry, S.I. (1989). Social and cultural effects on Indian learning 
style: Classroom implications. In B. Shade (Ed.), Culture, style and the educative 
process. Springfield, IL: Charles Thomas. 

Philips, S.U. (1983). The invisible culture: Communication in classroom and community 
on the Warm Springs Indian Reservation. New York: Longman. 

Saral, T. (1976, April 14-17). Consciousness theory and intercultural communication. 
Paper presented at the International Communication Association, Portland, OR. 

Shade, D.J. (1978). The social-psychological characteristics of achieving black 
children. Negro Educational Review, 29, 80-86. 

Shade, D.J. (1989). Culture style and the educative process. Springfield, IL: Charles 
C Thomas. (1976). 

Staples, R. Black culture and personality. In R. Staples (Ed.), Introduction to Black 
sociology. New York: McCraw Hill. 


Sue, D.W. (1981). Counseling the culturally different. New York: John Wiley 
and Sons. 

Trevarthen, C. (1988). Universal co-operative motives: How^ infants begin to know 
the language and culture of their parents. In G. Jahoda & I.M. Lewis (Eds.), 
Acquiring culture: Cross-cultural studies in child development. London: Croom Helm. 

Triandis, H. (1972). The analysis of subjective culture. New York: Wiley-Interscience. 

Trimble, J.E. (1981). Value differentials and their importance in counseling 
American Indians. In P. Pedersen, J. Draguns, W. Lonner, & J. Trimble (Eds.), 
Counseling across cultures: Revised and expanded edition, Honolulu, HI: East- 
West Center. 

Witkin, H.A. (1962). Psychological differentiation. New York: John Wiley and Sons. 




Communication and 

Community Participation in 

Program Evaluation Processes 

Robert T. Trotter II. Ph.D. 


Local communities play a vital role in the potential success or 
failure of culturally competent substance abuse programs. At the 
same time, community involvement generates evaluation issues 
that must be accommodated during various stages in the devel- 
opment, maintenance, change, or closure of community-based pro- 
grams (Randall-David, 1989; Nutting, 1990; Bracht, 1990; Whyte, 
1990). Representative community participation is particularly vi- 
tal to the success of evaluation plans for culturally competent sub- 
stance abuse prevention projects in ethnically and racially diverse 
communities. Setting up community-oriented evaluation designs, 
complemented by strong communication channels that work 
throughout the life of a program, can strengthen a program's po- 
tential for success, while lack of local participation can move it 
toward unfortunate and untimely demise. This chapter discusses 
some of the conditions that allow evaluators to work with, rather 
than against, the local community in setting program goals, iden- 
tifying process and outcome evaluation measures, and maintain- 
ing long-term community support for a program. 


There are several primary conditions that have an important 
impact on community-based evaluation plans. These include the 
cultural and sociopolitical environment within which the program 
and the evaluation process must coexist, the availability of local 
resources for conducting an evaluation, the existence of current 
methodological approaches that either support or hamper cultur- 
ally competent evaluation in a community context, and the con- 
ditions surrounding the communication of evaluation and research 
findings back to the community from which they are derived. 

The community's cultural and sociopolitical configuration cre- 
ates a general milieu for evaluation by establishing a framework 
to identify the local view of substance abuse problems and prob- 
lem solutions. This environment is inevitably pluralistic and may 
be polarized along significant dimensions that are locally impor- 
tant and even unique. This sociopolitical framework often identi- 
fies the areas of strongest local concern for measures of the success 
for substance abuse programs. This has an impact on evaluation 
strategies by potentially increasing or decreasing the credibility 
of evaluation findings, depending on whether these local priori- 
ties are measured for their impact by the evaluation process of 
the program. The results of a program evaluation, in turn, often 
have an impact on local politics inside and outside of the project, 
and can reconfirm or challenge the community's vision of its pri- 
orities for drug and alcohol abuse prevention programs. 

The local resources for evaluation, including project person- 
nel and community participation, can either enhance or inhibit 
the conduct of culturally competent evaluations. These resources 
dynamically interact with the evaluation strategies that are avail- 
able to community-based programs. Current methodological ap- 
proaches must often be adapted to local resources. The personnel 
skills that are available, and the fiscal resources that the program 
can generate to support evaluation goals, may not match ideal 
program evaluation needs. It is of little use to collect methodologi- 
cally elegant data that cannot be analyzed due to a lack of trained 
evaluators and computer facilities. And, in some cases, this type 
of data cannot be collected in the first place because of the inevi- 
table expense of elegance. This condition creates as many diffi- 
cult circumstances as methodologically incorrect data that cannot 


be defended. Neither of these problematic processes results in data 
that support broader program goals, nor can that type of data be 
safely disseminated. 

Finally, the public and semipublic dissemination mechanisms 
that provide information to different constituencies about project 
outcomes also play a key supporting role for program evaluation 
and maintenance. These and other parallel issues are explored in 
some detail below. 

General Community Concepts 
and Issues 

There are multiple definitions of community In this chapter, a 
community is a geographically bounded set of individuals, kin- 
ship units, and social institutions that interact in a culturally con- 
gruent manner to create and maintain potentially viable living 
conditions for human beings. This concept of a comraunity is simi- 
lar to that of a defined environmental unit, for analytical purposes. 
It must be redefined for each community that is to be described 
and evaluated, and it can be analyzed at multiple levels of inter- 
action. Although the concept of a global community has been pro- 
posed, the idea of community promoted here is less ambitious. It 
includes a small geographical area — one that is normally named 
by the individuals living in it, in which direct interaction among 
its members is possible. This is similar to the concept of commu- 
nity used in defining elements of community-oriented primary 
care programs (Strelnick, 1990), a useful construct for the devel- 
opment of evaluation components for community-based substance 
abuse prevention efforts. Under this definition, a metropolitan 
area could be considered a community, for one level of analysis. 
On the other hand, a geographically bounded neighborhood or 
any ethnic /racial, cultural, religious, or socioeconomic enclave 
could equally be defined and analyzed as a community. The con- 
cept of community contains the assumption that its members could 
individually or collectively interact with one another face to face, 
under the right circumstances, and would generally recognize each 
other 's membership, given the correct recognition information. 


The basic purpose of a community is to promote or reinforce 
the cultural norms of the group and to provide a framework for 
survival through all the processes supported by the community 
above the level of the individual and the family, such as general 
welfare of its members, protection from civil harm, mutual assis- 
tance through voluntary organizations, smooth economic trans- 
actions, and normal government functions, to name a few. This 
definition makes the concept of "community'' a useful evalua- 
tion and analytical construct. It suggests that there can be both 
individual and collective measurements of the success of a sub- 
stance abuse program. On an individual level, the change in a 
person's behavior from negative to positive (however defined) is 
the most common unit of measure. At the community level, 
changes in norms or a reduction in collective measures of nega- 
tive behavior (e.g., drug arrests, drunk-driving-related accidents) 
are possible. In fact, there appear to be more measures of commu- 
nity change available for evaluation of substance abuse preven- 
tion programs than there are measures of individual change. 

A community's cultural, racial, and socioeconomic circum- 
stances are all conditions in the local environment that affect the 
outcomes of abuse prevention programs; thus the community con- 
cept is useful in helping evaluators construct the most appropriate 
boundaries for their analysis, rather than trying to measure pro- 
gram impacts in social groups that would not have an opportunity 
to be affected. The community concept allows evaluators to accom- 
modate divergent conditions that are important in tailoring the 
abuse prevention program and its evaluation to clearly defined 
populations, value systems, and local circumstances. 

The identification of a community unit involves collecting all 
the appropriate parameters that define the boundaries and limits 
of the community, both from the view of the evaluation expert and 
the view of the members of that community. This often involves 
exploring the different ways that a community can be delineated, 
combined with the effects these definitions have on specific assess- 
ment strategies. The primary definitions normally use geographi- 
cal, demographic, cultural, ethnic, socioeconomic, physical, political, 
and environmental variables to construct community boundaries. 
Each combination of variables used has a somewhat different im- 
pact and may subtly determine at least some of the variables that 


are most appropriate for evaluating and interpreting program and 
individual outcomes. For example, if cultural orientation is used to 
define community boundaries, it is necessary to ensure that the 
language and the cultural meaning of the evaluation instruments 
match appropriately with the community, and that the evaluation 
be placed within an appropriate cultural framework, including 
measures of acculturation to the dominant and to the local culture. 
The actual evaluation measures that are appropriate for each of 
these circumstances are described in other chapters in this volume. 
The following sections identify the key conditions that need to 
be taken into account in constructing successful programs and their 
associated evaluation strategies in culturally diverse communities. 

Community Conditions 
TInat Affect Evaluation 

Substance use and abuse take place within a changeable, often 
negative or conflictive sociopolitical environment. There are sig- 
nificant differences in the ways that laws and regulations about 
what can be consumed and who can be involved in alcohol and 
other drug consumption are interpreted and enforced in different 
communities. Local values and cultural variation in the ways that 
alcohol and other drugs are viewed can have positive or negative 
consequences on substance abuse program outcomes and on 
evaluation strategies. Many of the negative, and the positive, im- 
pacts on community-based substance abuse prevention programs 
are the result of conditions that are external to the project, such as 
local law enforcement efforts, government regulations, and the 
availability of corollary services. These external forces must be 
identified and discussed in any evaluation plan. An assessment 
process that ignores them does so at its own peril. For example, if 
drinking and driving issues are fundamental emotional and po- 
litical elements in the local culture, then the success of a program 
may need to be measured, in part, in relation to its effect on the 
reduction of this problem, as well as other targeted conditions. If 
prescription drug abuse is a concern, then intervention programs 
that provide community feedback on the rehabilitation of people 
with medical addictions may be important. 


Evaluation strategies must also take into account local sensi- 
tivities. Many ethnic /racial communities in the United States are 
expressing alarm over the growing evidence that drug-related 
problems are not uniformly distributed. Substance abuse prob- 
lems appear to occur in higher levels in some ethnic communities 
as opposed to others, and this information has been used to stig- 
matize those communities. One result is that such communities 
are reluctant to agree to evaluation programs that would rein- 
force these negative views. Other epidemiological concerns are 
also crossing over from scientific to political concerns about sub- 
stance abuse program goals. The most highly stigmatized disease 
in the United States, at present, appears to be AIDS. Injection drug 
use and sex for drugs have been directly linked to the spread of 
HIV in the United States, and there appears to be a difference in 
the levels of drug-related HIV infection encountered in different 
regions of the country and in different ethnic populations. This 
type of double stigmatization of local cultural groups creates even 
more concern about community participation in substance abuse 
program evaluations. In some cases, the rejection of program 
evaluation results is strongly linked to avoiding ''more bad news'' 
for ethnic /racial populations. 

Each of these conditions has a direct impact on the design of 
the instruments intended to measure the effectiveness of a pro- 
gram, as sociopolitical "outcomes" of substance abuse programs 
are increasingly intertwined with epidemiological outcomes. Some 
of the strategies for incorporating these issues in program apprais- 
als are described by Stecher and Davis (1987) in their introduc- 
tory work on focusing evaluations at the correct level of analysis. 
Others are covered by King, Morris, and Fitz-Gibbon (1987). The 
following sections provide other issues that must also be taken 
into account. 

Nongovernmental Institutions and 
Community Programs 

There are a growing number of local interest groups and volun- 
tary organizations that target substance abuse control and reha- 
bilitation issues. Many of these organizations are grassroots or 


community based and often are developed in response to a per- 
ceived lack of government (local to Federal) competence in deal- 
ing with such problems. These groups include self-help 
organizations like Alcoholics Anonymous and Narcotics Anony- 
mous or the special lobbying groups like Mothers Against Drunk 
Driving. The presence or absence of these voluntary associations, 
and the level of intensity of their operations, can have a direct 
effect on the cultural ecology of the community They are com- 
monly single-issue driven, with a level of drive that is very high 
in some communities and virtually invisible in others. They tend 
to target narrow substance abuse or community outcomes, which 
may either reinforce or inhibit the goals of more generic abuse 
problem prevention and intervention programs through their 
singleness of purpose. The local effects of these movements need 
to be addressed in any comprehensive program evaluation plan. 
Many of these organizations are powerful or effective in some of 
the populations in a town but may have virtually no impact or no 
presence in other cultural, ethnic /racial, or socioeconomic groups. 
They can be particularly problematic where they do not meet the 
needs of a diverse community structure, favoring one socioeco- 
nomic or cultural group over others. 

The Media 

Many substance abuse prevention and intervention programs have 
some element built into them that stresses an impact on the com- 
munity through public awareness campaigns. This means that 
culturally competent evaluation strategies need to establish a base 
line evaluation of media messages that are related to program 
goals and then to evaluate the program's impact on the number, 
quality, and targeted message impact on the media for each "com- 
munity'' that is part of the evaluation plan. 

This evaluation strategy is complicated by having to deal with 
numerous mixed messages. Local and national media often ac- 
cept or publicly promote a general societal obligation to raise com- 
munity awareness of substance abuse problems and to promote 
the reduction of abuse risks at a community level. At the same time, 
advertisers have increasingly targeted women and ethnic /racial 


groups with highly effective cultural marketing programs that ap- 
pear to have increased alcohol and tobacco consumption. Thus, 
within this context, local community standards of "good taste" 
can encourage the media or prevent them from publicizing the 
full range of the impacts of drugs or alcohol on human behavior 
(positive or negative). Some key types of preventive or protective 
behaviors may be ignored through special interest pressure. There- 
fore, some substance abuse prevention programs tend to focus 
their media campaigns on impacts that do not carry a "moral'' 
loading effect, following the general U.S. precept of separating 
church and state. The mixed media messages (and "blind" areas 
where messages are not sent) in each community create a signifi- 
cant challenge for culturally competent evaluation of the impact 
of a public awareness campaign from a single program or even a 
group of programs promoting a single public message. 

Public Attitudes and Beiiefs 

Public beliefs and attitudes toward alcohol or drug use compose 
a critical set of baseline information that should be collected and 
later evaluated for substance abuse prevention or intervention 
programs. Much of this information should be collected during 
the initial needs assessments for the community and repeated at 
regular intervals to monitor change. It is extremely useful to know 
the knowledge, belief, and attitudes about abuse issues and pro- 
grams represented in the community before any attempt to change 
those conditions. Then, these baseline data can provide a poten- 
tial measure of the distance and the rate of change caused by a 
particular program on both individual and community variables. 
The baseline data must also be assessed against the potential con- 
founding effects of other, competing or complementary, substance 
abuse prevention programs that exist in the community. 

Stages of Community Involvement 

Governmental support of abuse programs, from the local to Fed- 
eral level, has created a boom-and-bust cycle that is the common 
denominator in ethnic /racial community substance abuse prob- 
lem prevention programs. New projects are announced, funded. 


and allowed to die on the basis of policies and priorities that are 
outside of the control of local "folks/' This condition is well known 
and is firmly embedded in any ethnic /racial community's re- 
sponse to new programs or changes in existing programs. There 
is a considerable amount of skepticism about the process. New 
programs are generally accepted from the viewpoint that some- 
thing is better than nothing, although this assumption has all too 
often proved to be wrong. 

If abuse prevention programs are to have a chance of being 
successfully adopted by communities, it is critical to provide a 
mechanism that results in real, rather than token, local-level in- 
volvement in the design, maintenance, and evaluation of the pro- 
grams (Chekki, 1979; Archer, Kelly & Bisch, 1984). Over the past 
20 years, there have been numerous government and regulatory 
attempts to promote ''grassroots'' or community involvement in 
federally subsidized programs. Local communities have often 
adopted these provisions, only to let them disappear along with 
the funding for the specific program that mandated them. Never- 
theless, local involvement, and more appropriately, local control 
of substance abuse programs are critical elements in the ethics 
and politics of abuse research, especially in conforming to the 
complex ethics of research and evaluation in racial /ethnic com- 
munities (Manson & Trimble, 1982; Beauvais & Trimble, 1991). 

All the different project development and maintenance phases 
can be the critical times when community involvement needs to 
be secured for the evaluation for a culturally competent program. 
These phases include pre-project planning, program design, 
startup of program activities, general program maintenance, and 
program closure. This sets the standard of evaluation of the com- 
plex dynamics that exist in a culturally diverse community. Each 
stage of program development can be identified as having inter- 
locking but somewhat differing assessment goals, objectives, and 
measurement needs. The development of the evaluation measures 
for each stage should include information about the probable com- 
munity reaction to and need for involvement in those processes 
(Orlandi, 1992). The following sections provide some details about 
these project phases and the evaluation and community partici- 
pation needs that accompany them. 


Pre-projecf Planning 

Substance abuse prevention programs are very often introduced 
into ethnic /racial communities from the outside. High-prestige 
universities (and not so prestigious ones) are famous for writing 
grants to test new prevention, evaluation, and intervention theo- 
ries and methods targeting racial, ethnic, or other underserved 
populations without having consulted and without having been 
asked by the local community to create the program. Some of these 
programs are based on data from community needs assessments, 
others on more general needs statements that are nevertheless 
highly persuasive. These programs, when funded (as they fre- 
quently are, since they are well designed from the perspective of 
the funding agency), are then introduced to the community or 
simply opened without prior consultation. This can cause serious 
design flaws and may create program evaluation disasters. 

Since the design and evaluation of these projects is directly 
affected by local circumstances, the ideal is to create a community 
partnership in the project from its earliest stages. New programs 
(and their attached evaluation components) need the acceptance 
of various leaders and influence groups in the community prior 
to their development. Too many programs are designed to ad- 
dress issues that are of low importance to the community (how- 
ever great their impact on prevention theory). Too many projects 
ignore the conditions that are of the highest priority for local citi- 
zens. Incorporation of these priorities in substance abuse programs 
can lead to strong support by the community This knowledge 
can then expedite the design of programs that meet both local 
and national goals, at many different levels. 

For example, the imposition of "outsider programs'' designed 
for Native- American groups in the United States has been so com- 
mon that most tribes have instituted local Institutional Review 
Boards whose primary purpose is to protect the community from 
additional research and demonstration projects that are irrelevant 
to community needs, regardless of their potential scientific merit. 
New substance abuse programs must now be reviewed and ac- 
cepted by local community boards, by the tribal government, and 
in some cases by Indian Health Service personnel, through for- 

mal review processes that vary from location to location. These 
new procedures make it imperative that individuals who are con- 
templating programs designed from the outside hold introduc- 
tory meetings with local officials prior to the development of the 
project. Tliese meetings are an excellent time to begin establish- 
ing the types of goals and objectives that are acceptable to the 
local community and can then lead to the development of highly 
relevant evaluation programs. This creates a mutual condition 
wherein it is in the best interest of the local community to allow 
testing additional scientific and national level priorities within 
the context of meeting some local goals. The combination tends 
to lead to a win-win condition for everyone involved. 

Project Design 

The second critical period for community involvement is during 
the project design stage. Incorporating community experts, to 
work in cooperation with substance abuse experts, can result in 
the development of a project that meets a large number of needs. 
The project design component must work out such local issues as 
a culturally appropriate governance structure, staffing that 
matches the cultural conditions extant in the community, and a 
culturally competent program design that is ''localized'' to meet 
existing conditions in that particular community. 

The days when there was sufficient funding to create substance 
abuse prevention and intervention programs that sounded inno- 
vative and attractive but could not scientifically prove that they 
have measurable positive consequences (let alone a measure of 
their cost-effectiveness) are probably gone forever. Most commu- 
nity leaders recognize the need for some type of measurement of 
program impact, both for their own political gain (by proving to 
the community that the projects they support have worked) and 
for the long-term viability of programs (by proving to outside 
funding sources that the projects are worth continuing when there 
are no local resources to continue them). 

Project development can be done within the context of com- 
munity involvement or by outsiders alone. Many communities 
have a strong interest in having outside groups provide the initial 


program development if there is not sufficient local expertise about 
new trends and programs. However, there are not any communi- 
ties that are comfortable with a program that is entirely imposed 
from the outside. Therefore, they request, or demand, a participa- 
tory development process that takes into account the division of 
expertise that is available. 

This development phase is the second point when project goals 
can be more clearly defined, potential impacts discussed by com- 
munity representatives, local priorities recognized, and the evalu- 
ation design modified to include local and theoretical and 
programmatic evaluation variables and conditions. The most com- 
mon local questions will revolve around the issue of, ''What's in 
it for us?" for both the community and the project personnel. 

Startup and Project Maintenance 

The early startup phase of a substance abuse prevention project is 
the time to reconfirm all evaluation strategies and to recheck the 
conditions that may produce external or intervening conditions 
that would affect the anticipated outcomes of the prevention or 
intervention program. Baseline data need to be collected as early 
as possible. It is also the best time to reaffirm local community 
priorities, which may or may not have changed between the time 
the project was originally contemplated and its actual initiation. 
This is also the time to either create, or to reactivate, local com- 
munity input through the use of community advisory boards, hir- 
ing of local experts, and incorporation of cultural experts in the 
evaluation process. These groups and conditions are described in 
more detail below. 

Project Close-Down 

Projects close down after either a brief or extended life. Minority 
communities are well aware of the boom-and-bust cycle of Fed- 
eral, State, and local funding. This makes them reluctant to accept 
experimental programs, since one effect of these programs is to 
raise people's expectations that something will be done about a 
local problem. All too often when an experimental model, or 
startup program, shows itself to be effective, funding is eliminated 


on the assumption that the local community will be able to con- 
tinue the program in the future. Since the conditions that made 
the community a target of these trial programs are often the same 
conditions that prevent it from maintaining programs through 
local resources, the close-down phase of a project often has seri- 
ous negative impacts on the community 

Closing pilot and demonstration programs is often unavoid- 
able, but it is very important that this condition be taken into ac- 
count in designing both the evaluation process and involving the 
community leadership in the project. This involvement can lessen 
the negative feelings left behind by the closing (where it is inevi- 
table) and may increase the probability of local support for the 
continuation of some or all of the program or its goals. If the pro- 
gram was unsuccessful, there is generally a need to explain its 
failure, based on concrete evaluation data. The involvement of 
local community leaders and members can help lessen the nega- 
tive or long-term consequences of that single program failure. 
Involving community members in a program's closing can also 
increase the probability that outsiders will have future opportu- 
nities to help create programs in the community rather than be- 
ing rejected out of hand because they were seen as having 
abandoned the community and avoiding a commitment to con- 
tinuing the program. 

Use of Local Resources 
for Program Evaluation 

The ultimate design of community-based evaluation components 
of a substance abuse program depends on several elements: the 
goals and purposes of the program, the availability of local ex- 
perts and program participants, public and private perceptions 
of the program, and the cultural parameters of abuse conditions 
in the local area. The design should also include the economics 
and politics of alcohol or drug use in the local area. As noted in 
the section above, all of these conditions make it important that 
the project foster local participation in the development and as- 
sessment of substance abuse prevention programs. The follow- 
ing sections provide an overview of the types of local expertise 


that are normally available as partners and feedback groups for 
evaluation programs, along with some of the areas of expertise 
that these groups bring to the project and its evaluation design. 

Types of Local Participants Available 

Programs designed to meet the needs of multicultural communi- 
ties normally have a number of individuals associated with them 
who can provide information that assists in meeting ongoing pro- 
gram development and evaluation requirements. These include 
the following groups of people. 

Community Leaders 

Community leaders can be local government officials; business, 
professional, and religious leaders; and nonprofit organization ex- 
ecutives. On the other hand, many recognized community leaders 
are locally credible individuals who have the well-being of the 
defined community as their first priority, but who do not participate 
in government or business affairs. Part of the challenge to culturally 
competent evaluation strategies is to identify all of the appropri- 
ate t3rpes of leaders to be represented in the evaluation process. 

These various kinds of leaders are important in several stages 
of project development (e.g., needs assessment, program design, 
monitoring program impact on the community, policy impacts of 
a program) and are most commonly incorporated into the project 
at a policy level, on community oversight boards. The basic de- 
sign of the program determines the types of local experts brought 
into this relationship. For example, government representatives 
are often necessary as participants in order to secure positive in- 
teractions with local agencies that have an impact on substance 
abuse problems, especially if the agencies are in any type of juris- 
dictional conflict, which is common. 

One of the important project functions of local experts is to iden- 
tify local priorities through their own knowledge of the commu- 
nity and their community contacts and networks. Community 
expectations can play an important role in communication and in- 
teraction with a particular program constituency, and these experts 
should be well aware of the priorities and the local mechanisms for 
setting up positive communications with appropriate groups. With- 


out their input, the priorities set for a program by the community 
may differ drastically from those set by the program staff and con- 
sultants. Identifying the similarities and differences in priorities for 
each contingency can be a factor in measuring either successful or 
unsuccessful program outcomes. These leaders are also a very im- 
portant element in protecting the program from political problems. 

Many people who are involved in substance abuse programs 
avoid getting deeply involved in politics, especially the politics 
of power, culture, economics, and ethnicity. Power politics goes 
against the egalitarian ideals of our society. This makes local poli- 
tics one area of expertise that is crucial for culturally competent 
evaluators. Substance abuse programs are surrounded by or em- 
bedded in political issues and processes that have a direct impact 
on the success or failure of programs. 

There are at least four types of politics that should be covered 
by local leaders or assessed by program evaluators for the protec- 
tion of the project. The first is the power relationships that exist 
within the program structure itself (office politics). The second is 
the set of relationships that exist between the program and the 
people who use its services. Often these relationships are affected 
by ethnic and socioeconomic politics in broad ways. The political 
environment that surrounds the project in the broader commu- 
nity is the third condition that must be addressed. And the final 
set of political relationships is the overall policy environment (from 
county to Nation) that influences the program from outside the 
community The project assessment plans will need to be adapted 
to these conditions in their design, execution, and dissemination 
of information within the project and between the project and 
outside constituencies. 

Cultural Experts 

Human beliefs, values, goals, and ideals vary significantly across 
cultural boundaries, leading to the need to develop culturally com- 
petent substance abuse programs and evaluation strategies. This, 
in turn, makes it essential to gain the participation of local cul- 
tural experts who can help the project develop strategies that re- 
flect and support the diversity found within the community, yet 
meet the requirements for scientifically valid evaluation. These 


experts can assist in helping the program match language, val- 
ues, and local knowledge about the problem to larger project goals. 
Cultural experts have a positive community impact through their 
relationship to the program. They also are important in review- 
ing the design of evaluation strategies, especially in terms of com- 
prehensive and culturally relevant evaluation instruments and 
insights into the interpretation of evaluation data. The other area 
in which their expertise is invaluable is providing help in dealing 
with local conventions regarding taboo subjects. 

All cultural systems have some form of taboo knowledge: in- 
formation that is wrong to share with outsiders, knowledge that 
is dangerous for outsiders to have, or knowledge that is consid- 
ered improper for other insiders to know. The danger of outsiders 
having taboo knowledge is twofold. In some cases the threat is to 
the insider group — the information can be used to socially harm 
or embarrass the group members or make them feel bad. In other 
cases, the knowledge is dangerous to outsiders — it is of such a 
sensitive nature that they are put in social or even physical dan- 
ger by knowing it. Each social system has built-in mechanisms 
for dealing with taboo knowledge, and they differ from group to 
group. By developing an evaluation system that includes signifi- 
cant participation from the community, these potentially danger- 
ous or embarrassing issues can be handled in a culturally 
competent manner, within the context of the evaluation process, 
without harmful effects to the community or the e valuators. Even 
programs that must deliberately deal with these areas (e.g., drugs, 
sex, domestic violence) can successfully evaluate program and 
community impacts if they are handled appropriately. 

Advisory Boards and Grassroots Involvement 

Advisory boards are becoming the most common source of com- 
munity involvement. They are normally composed of commu- 
nity leaders, cultural experts, program participants, consumers, 
and substance abuse specialists, and are customarily used to set 
policy and act as a communication bridge into various communi- 
ties and constituencies. The board can be an important source of 
information for the development of both process and outcome 
evaluation instruments and tools for projects, and is very useful 


in providing advice about policy issues relating to the evaluation 
of project goals from the point of view of users or potential par- 
ticipants, rather than the staff. 

'^ Hidden" Experts and Gate Keepers 

In most communities, and in ethnic /racial communities in par- 
ticular, some of the most influential individuals are not visible or 
discernible to outsiders. In some cases, communities of color de- 
liberately shield their true leaders from the mainstream to protect 
them from disrespect or attack. Yet these individuals can "make 
or break" a program in that community. They act as hidden evalu- 
ators and gate keepers. If possible, these individuals should be 
identified and included as a key element in any evaluation of the 
community impact of a program. These people are often hard to 
find and are most effectively approached through other community 
members, such as board members. But it is more than worth the 
effort to incorporate them in the long-term evaluation of the im- 
pact of the program and in the dissemination of information about 
the program back into the community, because of their influence. 

Securing Cooperation 
from file Community 

Substance abuse prevention programs in culturally diverse com- 
munities must be a cooperative endeavor. The local community 
impact, including the experts described above, can be positive or 
negative, depending on the relationships established by the pro- 
gram, and the program outcomes (Archer et al, 1984). At the most 
generic level, the community can be asked to identify the relative 
level of need for a particular program (baseline data to determine 
how well the need is or will be met by the program). It can estab- 
lish locally grounded expectations about potential program out- 
comes. And community involvement in program evaluation can 
help keep the evaluation (and the program) from incorrectly en- 
croaching on taboo subjects, or politically sensitive or volatile is- 
sues, or from encountering unnecessary obstacles through known 
problems in design and communication of evaluation strategies. 
Program communication and evaluation needs have to be 


understood in this sociopolitical context (Attneave, 1989). The is- 
sues that should be addressed through this cooperative process 
include the insider /outsider situation, local involvement and com- 
mitment, and some symbolic and semantic issues surrounding 
the relationships between local individuals, program staff mem- 
bers, and consultants brought in from the outside to support re- 
search and evaluation in culturally diverse communities. 

Outsiders Versus Insiders ("You ain't one of us") 

The sense of belonging and local identity that provide the defini- 
tion, boundaries, and norms for a community also create poten- 
tial opposition between insiders and outsiders. One program 
strategy is to incorporate sufficient local talent and commitment 
to either reduce or eliminate this area of potential conflict and to 
handle conflicts within the community, as well. Numerous projects 
have floundered or died on the basis of whether they are per- 
ceived as being imposed or being locally connected. Programs 
that were not created by recognized community members can 
often be the target of suspicion, regardless of either their intent or 
the value of the services or products produced by the project. This 
condition needs to be explored and taken into account in any 
evaluation strategy. The most common mechanism for improv- 
ing community participation in programs and their evaluation 
components is to thoroughly involve the local experts described 
above in all aspects of the program and its assessment, within the 
context of the conditions described below. 

Consultants and the ''Stranger Effect" 

Even though communities are often suspicious of the motives of 
outsiders, there is a countervailing attitude of respect for people 
with impressive credentials, national reputations, and associations 
with prestigious institutions. This condition can be used to the 
advantage of both the program evaluation and the community 
itself. Consultants and evaluation experts must establish local 
credibility, but once it is established, they can have a large impact 
on a community and a project. It is a case of hearing what ''strang- 
ers'' have to say with more intensity than what is said by local 
people, whom you have heard many times in the past. The 

elements that structure the acceptance of these experts can be very 
different, depending on the constituency evaluating the experts' 
appropriateness to the community Professional credentials may 
be meaningless in some contexts, where only cultural competence 
is recognized as important. Conversely, without the appropriate 
credentials, some individuals are not recognized as having the 
ability to make crucial judgments about a program. These issues 
must be balanced in relation to a program's needs and its evalua- 
tion design. A mixed strategy is often best: evaluation schemes 
should take advantage of the stranger effect, while supporting 
the development of local expertise. 

One of the important elements of the stranger effect is that 
people will often talk to these outsiders about issues that would 
be too politically divisive to discuss with local people. It is some- 
times easier for outsiders to set up a condition of confidentiality 
since they do not have an association vested in local political or 
social factions. They can more easily gain access to evaluation data 
that would place a local evaluator at either an advantage or dis- 
advantage in relation to other project personnel or the commu- 
nity. On the other hand, outsiders are normally kept from seeing 
and finding out about certain taboo subjects. The stranger effect 
works to the disadvantage of the evaluation process, and for these 
evaluation targets a trusted insider is needed to collect the infor- 
mation. Combined evaluation teams that take these opposing con- 
ditions into account can be particularly effective. 

The Language and Politics of 
Research Versus Evaluation 

Symbolic and semantic issues can have an important impact on 
the local perceptions of a project and may need to be taken into 
account in the design of culturally competent evaluation processes. 
For example, many of the terms used to designate the people who 
are the source of information for evaluating a project (e.g., research 
subjects, informants, respondents) have negative meanings in vari- 
ous community contexts. 'Tnformants,'' relating to substance 
abuse programs, may be synonymous with ''snitches.'' Subjects 
are people who are powerless and who are subject to conditions 
they may not like and do not control. Respondents are people 


who have no say in determining the way they reply; they simply 
respond to other people's questions, whether those questions re- 
late to them or are understandable. Program evaluation designs must 
work with, instead of against, these types of semantic and symbolic 

At another level, there may be a major symbolic difference 
between research and evaluation at the community level. Some 
groups have had negative experiences with past research endeav- 
ors, especially those in which they perceived that the investiga- 
tion had no benefit to the population being studied. In 
Native- American, Hispanic-American, and African- American 
communities, leaders often express the sentiment that they do 
not need anyone else coming in to ''put us under a microscope''; 
they need programs that work. In these communities, research 
has taken on the symbolism of oppression — something that is done 
to others, often without their consent. In some cases, evaluation 
has also taken on these negative connotations, but not as 

In most instances, evaluation is conducted in the context of 
trying to determine how beneficial a program or program ele- 
ment is, rather than assessing general community conditions. 
Research is thought of as primarily benefiting the researcher (who 
gets articles, promotions, and tenure, regardless of what happens 
to the community). Evaluation is thought of as providing infor- 
mation that can support good programs. There is a certain irony 
to this situation, since evaluation and research strategies are nor- 
mally indistinguishable in terms of design, instrumentation, and 
analytical frameworks. However, the more evaluation becomes 
indistinguishable from research, in both language and process, 
the more likely that the current advantage of calling program as- 
sessment ''evaluation" will disappear in the future. 

On the other hand, evaluation can also run afoul of commu- 
nity priorities. In these cases, there are local questions about the 
cost of evaluating a program. The money for the evaluation pro- 
cess may be viewed as a resource taken directly out of the pool of 
money that could be spent for services. The political attitude and 
forces of outside evaluators may dictate that something, however 
ineffective, is better than nothing for a community that has noth- 


ing and can pressure the community to develop minimal or inef- 
fective evaluation strategies. This creates a condition in which 
culturally competent evaluators must develop appropriate lan- 
guage and answers to questions about the purposes, relative 
worth, and the potential ''outsider'' uses of their endeavors, or 
their efforts may be destroyed. They must be able to defend evalu- 
ation as a community resource and service, rather than have it 
become a source of political difficulties. 

Basic Do's and Don'ts for Community Evaluation 

There are several rules or strategies that can help create an effec- 
tive and locally credible evaluation system. These include stay- 
ing neutral within the context of political and social 
factionalization, being useful but avoiding being used, and ''de- 
livering the goods" in ways that are locally valued. When the 
evaluation components of programs become involved in taking 
sides, their credibility is lost, even when the evaluation is accu- 
rate. There are numerous examples of local factions co-opting 
evaluators for their own purposes, to the detriment of the pro- 
gram. Even if no major divisions exist, individuals (either in the 
community or in the project) will commonly try to influence the 
design and outcome of assessment efforts to promote their own 
social or political careers. Others may see program evaluation as 
an opportunity to promote their interpersonal agendas by attempt- 
ing to uncover "improper conduct" in a project. They will aspire 
to redefine an evaluation as a witch hunt. These are processes 
that must be avoided from the outset through the design of politi- 
cally neutral evaluation programs. The antidote to most of such 
problems is to establish a clear understanding of the purposes of 
the evaluation program from project inception to its final conclu- 
sion. This allows the program and the evaluators to provide ev- 
eryone with a consistent and neutral description of the appropriate 
uses and the permissible conditions for use of the evaluation data. 
It creates a situation where the supervision of the evaluation pro- 
gram is maintained at the appropriate locus of control — for the 
benefit of the project, not individuals or factions. 


Communicating Evaluation and 
Program Research Findings 

The overall evaluation process is not complete until the informa- 
tion collected is communicated to the appropriate audience. Once 
the community-level concerns for the construction of substance 
abuse programs have been accommodated, the project will begin 
to produce significant amounts of evaluation information based 
on the appropriate methodological design. These data immedi- 
ately become extremely valuable to many groups. The data need 
to be returned to multiple constituencies for action, but in an ap- 
propriately controlled fashion. 

The purpose of most substance abuse evaluation strategies is 
to provide information that will make the program stronger and 
be better at serving client needs, more influential, and more clearly 
worthy of local support, plus any one of a large number of other 
purposes. This means that dissemination of the evaluation infor- 
mation should be incorporated into the design of the system from 
the very beginning of the project in order to avoid co-option or 
misinterpretation of the data. This need often creates a delicate 
balancing act and takes considerable forethought to be handled 
effectively, honestly, and constructively. 

Two of the most common concerns affecting local credibility 
of the evaluation are the avoidance of overselling the project and 
the problem of raising and then dashing community expectations. 
In many projects, there is some pressure on the evaluation pro- 
cess to promise more than a project can deliver, to the long-term 
detriment of the project. On the other hand, many communities 
have experience with changing local. State, and Federal priorities 
for programs and services. This creates skepticism, and it also cre- 
ates a problem of continually inflating and then destroying 
people's hopes for improvement in their community. Negative 
program evaluations can be interpreted as placing the commu- 
nity at further risk of the loss of resources. 

The following sections describe some of the processes, constitu- 
encies, and impacts that should be taken into account in the dis- 
semination of culturally competent program evaluation information. 


Ongoing Communication Processes 

A substance abuse problem prevention program that serves an 
ethnically diverse clientele should maintain consistent, ongoing 
communication channels to each of the groups present in the com- 
munity that are important to program success. This ensures early 
detection of potential problems (good process evaluation proce- 
dures), and helps ensure the long-term viability of a program by 
developing support based on mutually understood goals and 
defensible outcomes. This communication process is, in part, the 
responsibility of the evaluation team. Some of the types of com- 
munication that can result from good evaluation design include 
frequent updates for community leaders, ongoing media expo- 
sure, good communication mechanisms for getting information 
to local institutions concerned with substance abuse issues, feed- 
back to cultural leaders, and feedback to program clients. This 
type of design is also valuable for communicating on a regular 
basis with grassroots community boards and advisory groups. 
Strategies for these processes are covered in some detail by Morris, 
Fitz-Gibbon, and Freeman (1987). 

Clianneis and Audiences 

Most of the communication channels and audiences that are impor- 
tant to the normal functioning and use of evaluation information are 
obvious, once the program is designed. However, a basic study of 
the literature reveals that some of these commonsense entities are 
strangely ignored by projects, to their detriment. Therefore, it 
seems reasonable to summarize them as a potential checklist for 
projects that are developing or changing their evaluation strategies. 
A culturally competent project has many communication chan- 
nels open to it to provide information about the program to ap- 
propriate constituencies. These channels may vary locally and 
through cultural preference or the availability of resources to in- 
dividuals in these communities. They can also be mediated or 
constricted by language preferences and education levels. The 
common communication channels range in scale from commu- 
nity-wide channels (e.g., radio, television, newspaper) to more 
intimate or focused communication routes (e.g., newsletters, word 


of mouth, posters). The broadest include the locally available 
media, print or broadcast, and well-established institutional chan- 
nels that involve intergovernment and nonprofit agency boards, 
newsletters, and informal communication systems. 

The next level of communication is to make evaluation re- 
sults available to community participants, through board func- 
tions and targeted information campaigns (e.g., block programs, 
letters). Evaluation information can also be effectively dissemi- 
nated through public forums, such as local health fairs, educa- 
tional forums, and the like. Information can also be presented 
orally to such interest groups as local civic organizations. All of 
these assume that the project is providing consistent and frequent 
evaluation information to its own employees and board members. 

These channels further assume the existence of multiple au- 
diences for any evaluation findings: project personnel, clients, 
policy makers, the public, the media, and competing institutions. 
There is a dynamic tension created for most evaluation programs 
between the desire for maximum dissemination and a desire to 
control the way in that evaluation information will be interpreted. 
This creates a need for the project personnel to discuss the spe- 
cific types and volume of information that should go to each au- 
dience and through each communication channel. These 
discussions require evaluators to explore the ethics, politics, and 
logistics of providing both positive and negative evaluation in- 
formation to different groups, issues that are pragmatically ad- 
dressed by Morris et al. (1987) and Sieber (1991) with advice that 
relates directly back to the design and purpose of the evaluation 
program in the first place. 

Conclusions and Recommendations 

Accommodating the needs, issues, and conditions described in 
this chapter should be thought of as a cyclical rather than linear 
process. Most of the results of an evaluation program are pro- 
duced in a linear sequence, but the policy behind them, the audi- 
ence for them, and the ongoing relationships between the program 
and the community are a dynamic and evolutionary spiral that 
does not have a clear beginning and end, only cycles of interac- 


i*-^.i^ l^h'■l\ '"'- 

tion and change. The evaluation and communication efforts that 
Hnk a project to a community need to be evolutionary in order to 
accommodate the interactions implicit in both ongoing programs 
and continuing evaluation processes. Evaluation results and recom- 
mendations should be put into a repeating and changing format, 
rather than a linear checklist for completion. Each program ele- 
ment can cycle in and out of importance, according to what has 
been accomplished in the past and what is anticipated for the future. 
This condition creates an ongoing problem for program evalu- 
ators who either need to live in a linear world (based on the cul- 
ture of the program) or need to accommodate cycles of change 
within a product-oriented system. Viewing these issues from a 
nonlinear perspective creates an opportunity for designing evalu- 
ation programs that are culturally competent and that may be 
much more acceptable to the communities that need the types of 
workable substance abuse prevention programs described in this 


Archer, S.E., Kelly, CD., and Bisch, S.A., (1984). Implementing change in 
communities: A collaborative process. St. Louis, MO: C.V. Mosby Company. 

Attneave, C. (1989). Who has the responsibility? An evolving model to resolve 
ethical problems in intercultural research. American Indian Alaska Native Mental 
Health Research 2(3), 18-24. 

Beauvais, E, and Trimble, J.E. (1991). The role of the researcher in evaluating American 
Indian drug abuse prevention programs. Manuscript supported by the National 
Institute on Drug Abuse and National Institute on Alcohol Abuse and 

Bracht, N. (1990). Health promotion at the community level. Newbury Park, CA: 
Sage Publications. 

Chekki, D. A. (1979). Community development: Theory and method of planned change. 
New Delhi: Vikas Publishing House PVT LTD. 

King, J.A., Morris, L.L., & Fitz-Gibbon, C.T. (1987). How to assess program 
implementation. Newbury Park, CA: Sage Publications. 

Manson, S.M., & Trimble, J.E. (1982). American Indian and Alaskan Native 
communities: Past efforts, future inquiries. In L.R. Snowden (Ed.), Reaching 
the undeserved: Mental health needs of neglected populations (pp. 143-163). Beverly 
Hills, CA: Sage Publications. 


Morris, L.L., Fitz-Gibbon, C.T., & Freeman, M.E. (1987). How to communicate 
evaluation findings. Newbury Park, CA: Sage Publications. 

Nutting, P.A. (Ed.) (1990). Community oriented primary care: From principle to 
practice. Albuquerque, NM: University of New Mexico Press. 

Orlandi, M.A. (1992). The challenge of evaluating community based prevention 
programs: A cross-cultural perspective. In M.A. Orlandi (Ed.), Cultural 
competence for evaluators: A guide for alcohol and other drug abuse prevention 
practitioners working with ethnic/racial communities (DHHS Publication No. 
(ADM) 92-60067). Washington, DC: Alcohol, Drug Abuse, and Mental Health 

Randall-David, E. (1989). Strategies for working with culturally diverse communities 
and clients. Washington, DC: Office of Maternal and Child Health, Bureau of 
Maternal and Child Health Resources Development, U.S. Department of 
Health and Human Services. 

Sieber, J.E. (1991). Sharing social science data: Advantages and challenges. Newbury 
Park, CA: Sage Publications. 

Stecher, B.M., & Davis, WA. (1987). How to focus an evaluation. Newbury Park, 
CA: Sage Publications. 

Strelnick, A.H. (1990). The community-defining process in community oriented 
primary care. In P.A. Nutting (Ed.), Community oriented primary care: From 
principle to practice. Albuquerque, NM: University of New Mexico Press. 

Whyte, W.F. (1990). Participatory action research. Newbury Park, CA: Sage 




...WC^ .^^^^ ..WO^ 



4,a^^^ Pte^^' t>fevei^^ 







0^' _^^,ci6^ 

.et^o^ ';-^"" ^^^^^^ .^ .^aYt^ 





• OK^ 






,.eve-^^^^>^ .a^^e^^"' 


..^^^ ..O^KSI 



.et-- ^ ^otv^ 





.WORE- .Avvt«N5^- ,.oR^' WO^' 

,Va^'"^ ttev 




,,...*"^^^ „....'»"'^:ss^' ,„.-"*''>— *"^^' 

• ' -rt t ^* ' " -=^^^ 





.'OfiSS! _ ,,,05^! 


etitio^ 'j.^^^ ^ _,.^titio^ ^ >--'- ^^^^io^ 












DHHS PubHcation No. (SMA)96-3110 
Printed 1996